University of Colorado Health at Memorial Hospital v. Burwell

• Court: District Court, District of Columbia
• Decided: June 17, 2022
• Docket: Civil Action No. 2014-1220
• Status: Published

Opinion

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

UNIVERSITY OF COLORADO HEALTH : AT MEMORIAL HOSPITAL, et al., : : Plaintiffs, : Civil Action No.: 14-1220 (RC) : v. : Re Document Nos.: 184, 185, 188 : XAVIER BECERRA, Secretary of : Health and Human Services, : : Defendant. :

MEMORANDUM OPINION

GRANTING IN PART AND DENYING IN PART PLAINTIFFS’ MOTION FOR SUMMARY JUDGMENT; GRANTING IN PART AND DENYING IN PART DEFENDANT’S CROSS- MOTION FOR SUMMARY JUDGMENT; AND GRANTING IN PART AND DENYING IN PART PLAINTIFFS’ MOTION TO COMPLETE ADMINISTRATIVE RECORDS

I. INTRODUCTION

In advance of each fiscal year (“FY”), the Secretary of the Department of Health and

Human Services (“HHS” or the “Secretary”) engages in a notice-and-comment rulemaking

process to establish a number that will play a significant part in determining the extent to which

hospitals will receive Medicare reimbursement payments for certain extraordinarily costly

services performed during the fiscal year. Plaintiffs, a group of hospitals, ask the Court to vacate

the rules for FYs 2007–2013 because of alleged procedural and/or substantive defects in HHS’s

rulemaking proceedings for these years. The Court holds that the Secretary’s explanations of

certain decisions reached in the FY 2012 and FY 2013 rules were inadequate under the

Administrative Procedure Act; accordingly, it remands these rules to the Secretary for further explanation. Otherwise, the Court holds that the Secretary acted lawfully in promulgating the

remaining challenged rules.

II. BACKGROUND

A. Regulatory Framework

The Court assumes familiarity with its detailed descriptions of the regulations governing

the Medicare outlier payments program found in prior opinions in this case. See Mem. Op.

Granting in Part and Denying in Part Def.’s Partial Mot. Dismiss and Granting in Part and

Denying in Part Pls.’ Mot. Suppl. Admin. R. (“Mot. Dismiss Op.”), ECF No. 155; Mem. Op.

Granting Def.’s Mot. Leave to Suppl. Answer (“Mem. Op. Suppl.”), ECF No. 89; Mem. Op.

Granting Def.’s Mot. for Clarification (“Clarification Op.”), ECF No. 57; Mem. Op. Granting in

Part and Denying in Part Pls.’ Mot. to Compel Prod. of Complete Admin. R. (“Suppl. Rec.

Op.”), ECF No. 47. And it will provide additional detail as necessary throughout its analysis.

Still, for orientation, the Court directly repeats, with some modifications, part of the background

it provided in its most recent opinion in this case. Mot. Dismiss Op. at 1–9.

Under Medicare, the federal government reimburses hospitals for supplying medical

services to the elderly and disabled. See Social Security Amendments of 1965 (“Medicare Act”),

Pub. L. No. 89–97, tit. XVIII, 79 Stat. 286, 291.1 Providers are not reimbursed for the full costs

that they incur; instead, they are paid at fixed rates for different categories of services and

treatments, known as “diagnosis-related groups” (“DRGs”). See Billings Clinic v. Azar, 901

F.3d 301, 303 (D.C. Cir. 2018) (citation omitted). However, hospitals are also eligible for

certain outlier payments as a form of protection against unusually complicated and costly cases.

Id. at 304 (citing 42 U.S.C. § 1395ww(d)(5)(A)(ii)). These payments become available when the

1 Codified as amended in 42 U.S.C. § 1395 et seq.

2 provider’s (1) “cost-adjusted charges” for a case exceed (2) the sum of (2a) the default

reimbursement payment and (2b) a fixed dollar amount (known as the “outlier threshold” or the

“fixed loss threshold” (FLT) and determined by the Secretary through an annual rulemaking

process). Id. at 304 (citation omitted).

That first figure—the provider’s “cost-adjusted charges”—is intended to estimate the

provider’s real cost of care, without any markups, and is calculated by multiplying a provider’s

actual charges by a historical “cost-to-charge ratio.” Id. at 304–05 (citation omitted). The

second figure—the sum of the base reimbursement plus the fixed loss threshold—is known as

the “fixed-loss cost threshold.” Id. at 304 (citation omitted). Cost-adjusted charges above the

fixed-loss cost threshold are reimbursed at a rate intended to approximate the marginal cost of

care, currently set at 80 percent in most cases. Id. at 305 (citation omitted).

As an example: imagine a hospital charges $100,000 for an unusually complicated

procedure.2 The $100,000 will be multiplied by a cost-to-charge ratio (“CCR”) (imagine it’s

72:100 or 72 percent, which HHS will have calculated based on historical data), leaving $72,000

of cost-adjusted charges. Imagine too that the standard DRG reimbursement rate for this kind of

procedure is $8,000, and the fixed loss threshold set by the Secretary that year is $11,000. The

hospital will automatically receive the base reimbursement of $8,000. And because the cost-

adjusted charges ($72,000) are greater than the fixed-loss cost threshold ($19,000), the hospital is

also eligible for an outlier payment. That payment will be 80 percent of the difference between

the cost-adjusted charges ($72,000) and the fixed-loss cost threshold ($19,000), or $42,400.

2 This is based on example offered in the Secretary’s opening motion-to-dismiss brief, see Def.’s Mem. Supp. Mot. Dismiss at 6, ECF No. 139-1, which is in turn drawn from an August 29, 1997, Federal Register notice: Changes to the Hospital Inpatient Prospective Payment Systems and Fiscal Year 1998 Rates, 62 Fed. Reg. 45,966, 46,011 (Aug. 29, 1997).

3 Notice that when the fixed loss threshold is smaller, it is more likely that a hospital will receive

an outlier payment and that any outlier payment received will be greater.

That leaves an important question: how does the Secretary determine each fiscal year’s

fixed loss threshold? Well, Congress has limited the aggregate amount of Medicare outlier

payments to a narrow range: it “may not be less than 5 percent nor more than 6 percent of the

total payments projected or estimated to be made based on DRG prospective payment rates for

discharges in that year.” 42 U.S.C. § 1395ww(d)(5)(A)(iv). To satisfy this directive, HHS

conducts an annual rulemaking to set the fixed loss threshold at a level that it estimates will

result in total payments within the statutorily-determined range (more on that later). See Billings

Clinic, 901 F.3d at 306–07 (citation omitted). Specifically, since 1989, HHS has attempted to set

an annual threshold that will result in total outlier payments being 5.1 percent of all Medicare

payments. Id. at 307. Crucial to the Secretary’s projections are the providers’ estimated future

cost-to-charge ratios. Id. For instance, if HHS overestimates a future year’s cost-to-charge

ratios (expecting, say, 90 percent when it turns out to be 72 percent), then reimbursable, cost-

adjusted charges will be lower than expected—meaning that HHS may have set the fixed loss

threshold too high and therefore be at risk of undershooting its 5.1 percent payment target.

This is all the more important because, in order to fund outlier payments, the Secretary

withholds the predicted 5.1 percent from all other standard reimbursements. See 42 U.S.C. §

1395ww(d)(3)(B). And the Secretary need not take corrective action when the actual outlier

payments differ from the 5.1 percent target. See Dist. Hosp. Partners L.P. v. Burwell, 786 F.3d

46, 51 (D.C. Cir. 2015) (citing County of Los Angeles v. Shalala, 192 F.3d 1005, 1020 (D.C. Cir.

1999)). As a result, undershooting the 5.1 percent target results in a net loss of payments to

providers as a whole.

4 Procedurally, healthcare providers are reimbursed on a rolling basis, but at the end of

their fiscal years, they submit annual cost reports to so-called “medicare administrative

contractors” or “fiscal intermediaries.”3 See 42 U.S.C. § 1395h(a); 42 U.S.C. § 1395kk-1(a)(3)–

(a)(4)(B); 42 C.F.R. § 413.20(b). Fiscal intermediaries then issue a total reimbursement

determination for the entire year4 through a Notice of Program Reimbursement (“NPR”). 42

C.F.R. § 405.1803(a). Hospitals are permitted to challenge an NPR by appealing to the Provider

Reimbursement Review Board (“PRRB”), a specialized administrative body. 42 U.S.C. §

1395oo(a). Hospitals can in turn seek judicial review of a PRRB’s final decision. §

1395oo(f)(1). Providers also “have the right to obtain judicial review of any action of the fiscal

intermediary which involves a question of law or regulations relevant to the matters in

controversy whenever the [PRRB] determines . . . that it is without authority to decide the

question”; such determinations for expedited review can be made sua sponte by the PRRB or at

the request of a provider. Id. In either case, a district court reviews the challenged action

“pursuant to the applicable provisions” of the Administrative Procedure Act (“APA”). Id.

One other feature of the process bears mentioning at this stage. In the early 2000s, the

Secretary would determine a hospital’s cost-to-charge ratio using “cost and charge data from the

‘latest available settled cost report’ without any forward projections.” Billings Clinic, 901 F.3d

at 305. But this approach proved problematic, because

cost reports take several years to settle. And that time lag generated opportunities for abuse. Hospitals could manipulate their outlier payments by inflating current

3 “Medicare administrative contractor” is the current statutory terminology. See 42 U.S.C. § 1395h(a). Fiscal intermediary is an older term, see Palisades Gen. Hosp. Inc. v. Leavitt, 426 F.3d 400, 401 (D.C. Cir. 2005), but it remains in usage, see, e.g., 42 U.S.C. § 1395oo(a)(1)(A)(i). The Court will use both terms interchangeably to refer to the kind of entities described in 42 U.S.C. § 1395h(a). 4 Note that a hospital’s fiscal year may not align with the federal fiscal year, meaning that a single NPR may be governed by two different fiscal year thresholds.

5 charges so that the historic cost-to-charge ratio employed to calculate outlier payments did not reflect the hospital’s true costs. In those situations, the hospital’s cost-to-charge ratio would overstate actual costs, resulting in an inflated cost estimate for the current year’s claims.

Id. This trick came to be known as “turbo-charging.” Id. at 306. The Secretary responded to the

turbo-charging problem in 2003 by enacting a series of reforms, including “reserv[ing] the right

to recalculate a hospital’s eligibility [for an outlier payment] using actual cost data at the time of

settlement. Through this process, known as reconciliation, the agency [can] claw-back undue

outlier payments.” Id.(citations omitted).

B. Procedural History

Many of the plaintiff hospitals here were plaintiffs in two other related cases. Banner

Health v. Azar, No. 10-cv-1638 (D.D.C.) was filed in 2010. In addition to advancing some other

claims, the Banner Health plaintiffs challenged the fixed loss threshold determinations for

federal fiscal years 1997 through 2007. Banner Health v. Burwell, 126 F. Supp. 3d 28, 43

(D.D.C. 2015). The district court disposed of the plaintiffs’ claims through various motions to

dismiss and for summary judgment. See Banner Health v. Burwell, 174 F. Supp. 3d 206, 207–08

(D.D.C. 2016). The Circuit largely affirmed, though it reversed the district court’s grant of

summary judgment as to fiscal years 2004 through 2006 on the grounds that HHS inadequately

explained certain aspects of those threshold calculations. See Banner Health v. Price, 867 F.3d

1323, 1337–39 (D.C. Cir. 2017). The parties stipulated to dismissal of that case with prejudice

in June 2020. Order, Banner Health v. Azar, No. 10-cv-1638 (D.D.C. June 18, 2020).

Another group of cases were filed in 2013 and 2014 and were consolidated in Lee

Memorial Hospital v. Burwell, No. 13-cv-643 (D.D.C.). The Lee Memorial plaintiffs challenged

certain rulemaking actions taken in 2003 and the fixed loss threshold determinations for federal

fiscal years 2008 through 2011. The court granted summary judgment for the Secretary on all

6 the plaintiffs’ claims. Lee Mem’l Health Sys. v. Burwell, 206 F. Supp. 3d 307, 336 (D.D.C.

2016). On appeal of those cases (under the caption Billings Clinic v. Azar), the Circuit affirmed,

finding that the calculations were reasonable and that the challenge to the 2003 rulemaking

actions was precluded by Banner Health. See Billings Clinic, 901 F.3d at 302–03.

This case, University of Colorado Health at Memorial Hospital v. Azar, No. 14-cv-1220,

was consolidated for all purposes with seven later-filed cases. See Dec. 19, 2018 Order at 1,

ECF No. 108; Feb. 15, 2019 Order at 1, ECF No. 112; April 1, 2019 Order at 2, ECF No. 131.

The consolidated action comprises eight currently operative complaints from hospitals who

administratively appealed their cost reports from various years on the ground that their payments

on outlier claims were too low. See Pls.’ Mot. Summ. J. & Mem. Supp. at 10 (“Pls.’ Summ. J.

Mem.”), ECF No. 185. Each plaintiff hospital received from the PRRB a grant of expedited

judicial review regarding the validity of the fixed loss threshold regulations governing their cost

reports. See id. This Court dismissed certain claims as voluntarily abandoned and others as

barred by claim preclusion. See id.; Mot. Dismiss Op. at 14, 19.

C. The Instant Motions

The parties have filed cross motions for summary judgment in their favor on all

remaining claims. Pls.’ Summ. J. Mem.; Def.’ Cross Mot. Summ. J., ECF No. 188. In general,

Plaintiffs’ remaining claims assert that HHS failed to follow proper procedures and/or acted

unreasonably when it calculated fixed loss thresholds in rulemakings for fiscal years 2007–2016.

These errors allegedly caused HHS to set the fixed loss threshold too high for these years, which

in turn caused Plaintiffs to receive less in outlier payments than they otherwise would have.

Recall that each year, HHS “must estimate the number of outlier cases for the upcoming

year and set a threshold that it believes will result in outlier payments of 5.1%” of total Medicare

payments for the year. Billings Clinic, 901 F.3d at 307. To accomplish this, HHS “forecast[s]

7 the total outlier payments that it would make under various potential thresholds, compared to

total DRG payments it would make during the upcoming [fiscal year], until it finds a threshold

projected to produce total outlier payments at its 5.1% target.” Pls.’ Summ. J. Mem. at 5.

Generally, during the years at issue, HHS used the following method:

1. Assume patient cases in the coming FY will be the same as those in the MedPAR5 from two FYs before (e.g., for FY 2008, HHS used FY 2006 cases). 2. Simulate DRG-related payments on those cases using payment rates and policies for the coming FY. 3. Simulate any outlier payments on those cases, using the rules governing outlier payments and the following process: a. HHS increased the charges from the MedPAR file by two years’ worth of charge inflation, assuming charges would increase the same amount annually as the year-on-year increase reflected in that file. b. HHS then multiplied these forecasted charges by forecasted CCRs. To derive forecasted CCRs, HHS took each hospital’s most recently available CCR in the March [Provider Specific File (“PSF”)]. HHS also applied an “adjustment factor” to the CCRs, which was necessary because (1) the CCRs in the PSF would later be updated at least once, and in many cases twice, before their use to calculate individual outlier claims during the upcoming FY and (2) HHS had long recognized that CCRs have generally been decreasing over time. Thus, if CMS did not apply an accurate adjustment to the CCRs, it would systematically over-project hospital costs and outlier payments and set excessive thresholds.

Id. at 5–6 (citations and footnotes omitted); see also Billings Clinic, 901 F.3d at 307–08. HHS

continued to apply this method across the FY 2007–2013 threshold-setting rules. See Pls.’

Summ. J. Mem. at 20 n.13 (collecting sources).

Plaintiffs’ claims in this case relate to HHS’s calculation of the adjustment factor applied

to estimate CCRs for the coming year. The FY 2007 rule was the first in which HHS applied an

adjustment factor to forecast CCRs—in previous years, it had simply (at a general level) “take[n]

5 “MedPAR is a database that aggregates the claims submitted by hospitals to HHS.” Banner Health v. Price, 867 F.3d 1323, 1342 (D.C. Cir. 2017).

8 the historical cost-to-charge ratio from the most recent year available and project[ed] those

figures forward.” Billings Clinic, 901 F.3d at 307. The adjustment factor used in the FY 2007

rule consisted of two parts, one for each component of the cost-to-charge ratio: an estimate of

inflation of charges and an estimate of inflation of costs. The “charge inflation factor” consisted

simply of the average annual rate of change in charges per case over the past two years. Id. at

308; 71 Fed. Reg. 47,870,48,149 (Aug. 18, 2006). The “cost inflation factor” was “more

complex,” and “factored in both hospital-specific cost inflation and general inflation as measured

by the change in a standard market basket of goods and services.” Billings Clinic, 901 F.3d at

308. Thus, “HHS calculated . . . the average increase (over all relevant hospitals) in hospital

operating costs; and it divided that figure by the increase in the ‘market basket’ . . . for the same

time period.” Pls.’ Summ. J. Mem. at 16 (citing 71 Fed. Reg. at 48,150). “HHS then averaged

that number over three prior years.” Id.; see Billings Clinic, 901 F.3d at 308; 71 Fed. Reg. at

48,150.

Plaintiffs claim that this method of projecting cost inflation is both procedurally and

substantively unsound. They say that the Secretary violated the Medicare Act’s procedural

notice and comment requirements, see 42 U.S.C. § 1395hh(b)(1), by failing to announce the new

complex CCR adjustment method in the proposed rule for FY 2007 before implementing it in the

final FY 2007 rule and by failing to publicize certain documents related to the method during the

2007–2013 rulemakings. Pls.’ Summ. J. Mem. at 15–18, 20–21. Plaintiffs also claim that the

FY 2008 rule failed to comply with notice and comment procedures because the final rule

forecasted an increase in CCRs even though the proposed rule estimated a decrease in CCRs. Id.

at 18–19. As for substance, Plaintiffs assert that the Secretary was arbitrary and capricious in

setting the thresholds in various fiscal years because its selection of a three-year averaging period

9 for the cost inflation factor was arbitrary, because it persisted in using the same method even

though its selected outlier thresholds frequently failed to hit the intended 5.1 percent target,

because it did not sufficiently engage with commenters’ alternative suggestion of adjusting its

projected payments to account for its historical rate of underestimation, and because it did not

account in its projections for the possibility that it would recoup outlier payments during the

reconciliation process. Id. at 22, 27, 32–33.

In his cross motion for summary judgment and opposition to Plaintiffs’ motion, the

Secretary contends that each of Plaintiffs’ complaints about the threshold-setting rules is legally

deficient. Mem. Supp. Sec’y’s Cross Mot. Summ. J. & Opp’n Pls.’ Mot. Summ. J. at (“Def.’s

Summ. J. Mem.”), ECF No. 188-1 at 12–30. He also says certain hospitals’ claims are untimely

and/or barred by issue preclusion. Id. at 40–43.

On the same day Plaintiffs moved for summary judgment, they also moved to complete

the administrative records for the FY 2008–2013 rules with two documents purportedly showing

analysis relevant to HHS’s method for calculating CCR adjustment factors. Pls.’ Mot. Complete

Admin. R. at 1, ECF No. 184. Defendant opposes this motion. Sec’y’s Opp’n Pls.’ Mot.

Contesting Admin. R. at 1 (“Def.’s Admin. R. Opp’n”), ECF No. 186.

III. LEGAL STANDARDS

In a typical case, the Court must grant summary judgment to a movant who “shows that

there is no genuine dispute as to any material fact and the movant is entitled to judgment as a

matter of law.” Fed. R. Civ. P. 56(a); see also Winston & Strawn, LLP v. McLean, 843 F.3d 503,

505 (D.C. Cir. 2016). But in the context of the APA, the Court’s review of the administrative

record is limited. Sierra Club v. Mainella, 459 F. Supp. 2d 76, 89 (D.D.C. 2006) (citing Nat’l

Wilderness Inst. v. U.S. Army Corps of Eng’rs, 2005 WL 691775, *7 (D.D.C. Mar. 23, 2005)). It

10 is the agency’s role to resolve issues of fact and regulate in accordance with those facts. See

Sierra Club, 459 F. Supp. 2d at 90.The district court’s review is confined to determining

whether, as a matter of law, the evidence in the administrative record supports the agency’s

decision. Citizens for Resp. & Ethics in Washington v. SEC, 916 F. Supp. 2d 141, 145 (D.D.C.

2013). “Summary judgment thus serves as the mechanism for deciding, as a matter of law,

whether the agency action is supported by the administrative record and otherwise consistent

with the APA standard of review.” Id. (citing Richards v. INS, 554 F.2d 1173, 1177 & n.28

(D.C. Cir. 1977)).

As the foregoing standard suggests, in an APA case, a reviewing court normally “should

have before it neither more nor less information than did the agency when it made its decision.”

IMS, P.C. v. Alvarez, 129 F.3d 618, 623 (D.C. Cir. 1997) (quoting Walter O. Boswell Mem’l

Hosp. v. Heckler, 749 F.2d 788, 792 (D.C. Cir. 1984)). Agencies bear the responsibility of

compiling the administrative record, which must include all of the information that the agency

considered “either directly or indirectly.” Marcum v. Salazar, 751 F. Supp. 2d 74, 78 (D.D.C.

2010). The record that an agency produces “is entitled to a strong presumption of regularity.”

Id.

A party may seek to supplement the record produced by the agency, however, in “one of

two ways.” WildEarth Guardians v. Salazar, 670 F. Supp. 2d 1, 5 n.4 (D.D.C. 2009). First, a

party may seek to include “evidence that should have been properly a part of the administrative

record but was excluded by the agency.” Id. Where a plaintiff follows this first route,

supplementation is appropriate if the agency “did not include materials that were part of its

record, whether by design or accident . . . .” Marcum, 751 F. Supp. 2d at 78. But to overcome

11 the presumption of regularity, “a plaintiff must put forth concrete evidence that the documents it

seeks to ‘add’ to the record were actually before the decisionmakers.” Id.

Alternatively, a party may seek to supplement the record with “extra-judicial evidence

that was not initially before the agency but [which] the party believes should nonetheless be

included in the administrative record.” WildEarth Guardians, 670 F. Supp. 2d at 5 n.4. In these

circumstances, a more stringent standard applies. To “justify[ ] a departure from [the] general

rule” that review “is to be based on the full administrative record that was before the Secretary at

the time he made his decision,” a party must demonstrate one of three “unusual circumstances.”

Am. Wildlands v. Kempthorne, 530 F.3d 991, 1002 (D.C. Cir. 2008) (internal quotation marks

omitted). Those circumstances include: (1) when “the agency ‘deliberately or negligently

excluded documents that may have been adverse to its decision,’” (2) when “background

information [is] needed ‘to determine whether the agency considered all the relevant factors,’”

and (3) when “the ‘agency failed to explain administrative action so as to frustrate judicial

review.’” City of Dania Beach v. F.A.A., 628 F.3d 581, 590 (D.C. Cir. 2010) (quoting Am.

Wildlands, 530 F.3d at 1002).

IV. ANALYSIS: PROCEDURAL CHALLENGES

“Pursuant to the Medicare Act, 42 U.S.C. § 1395oo(f)(1), this Court reviews the

Secretary’s action under the familiar provisions of the APA, 5 U.S.C. § 706(2)(A).” Shands

Jacksonville Med. Ctr. v. Burwell, 139 F. Supp. 3d 240, 261 (D.D.C. 2015). Thus, HHS “must []

provide the public with a meaningful opportunity to comment on a proposed rule and must offer

reasoned responses to significant comments.” Id. Additionally, “[a]n agency may promulgate a

rule that differs from a proposed rule only if the final rule is a ‘logical outgrowth’ of the

proposed rule. A final rule is a logical outgrowth if affected parties should have anticipated that

12 the relevant modification was possible.” Allina Health Servs. v. Sebelius, 746 F.3d 1102, 1107

(D.C. Cir. 2014) (citations omitted); see Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158,

174 (2007) (noting that the logical outgrowth doctrine is an interpretation of the APA’s

requirement that a notice of proposed rulemaking contain “either the terms or substance of the

proposed rule or a description of the subjects and issues involved.” (quoting 5 U.S.C.

§ 553(b)(3))).

A. FY 2007 Rule

Plaintiffs raise two distinct yet related notice-and-comment arguments about the FY 2007

rule—they argue that HHS was required to present its CCR adjustment factor method for

comment before adopting that method in the FY 2007 final rule, and that HHS “withheld a

critical analysis that it had conducted to develop the method” in violation of the D.C. Circuit’s

Portland Cement doctrine, also known as the critical material doctrine. Pls.’ Summ. J. Mem. at

13, 16, 18 (citing the D.C. Circuit’s quotation in Am. Radio Relay League, Inc. v. F.C.C., 524

F.3d 227, 237 (D.C. Cir. 2008) of the standard articulated in Portland Cement Ass’n v.

Ruckelshaus, 486 F.2d 375, 393 (D.C. Cir. 1973), that “[i]t is not consonant with the purpose of a

rule-making proceeding to promulgate rules on the basis of . . . data that, [to a] critical degree, is

known only to the agency”). Neither objection persuades the Court.

1. The final FY 2007 rule was a logical outgrowth of the proposed FY 2007 rule.

In order to evaluate Plaintiffs’ logical outgrowth challenge, it is necessary to maintain a

bit of perspective. This Court previously held that the rules Plaintiffs challenge are “the overall

calculation of a given year’s fixed los[s] threshold”; that is, Plaintiffs did not preserve any

challenge “to the cost-to-charge ratio methodologies as . . . standalone” rules requiring their own

notice-and-comment proceedings. Mot. Dismiss Op. at 25. Therefore, the question is whether

the overall calculation of the fixed loss threshold implemented in the final FY 2007 Rule was a

13 logical outgrowth of the overall calculation of the fixed loss threshold proposed in the

Secretary’s notice of proposed rulemaking for FY 2007. The method of projecting CCRs is just

one element of this overall calculation, albeit a “crucial” one. Id.

In the notice of proposed rulemaking, the Secretary proposed using the same recent-

history-based method for projecting CCRs it had used up to that point:

As we have done in the past, we are proposing to establish the proposed FY 2007 outlier threshold using hospital cost-to charge ratios from the December 2005 update to the Provider-Specific File—the most recent available at the time of this proposed rule. This file includes cost-to-charge ratios that reflect implementation of the changes to the policy for determining the applicable cost-to charge ratios that became effective August 8, 2003 (68 FR 34494). Using this methodology, we are proposing to establish an outlier fixed-loss cost threshold for FY 2007 equal to the prospective payment rate for the DRG, plus any IME and DSH payments, and any addon payments for new technology, plus $25,530.

71 Fed. Reg. 23,996, 24,150 (Apr. 25, 2006); Administrative Record (“AR”) (FY 2007) at

012369. During the comment period, commenters expressed concern that the proposed threshold

was too high and would cause HHS to undershoot its 5.1 percent target. They “recommended

[that HHS] further refin[e] the outlier methodology,” including by “us[ing] an adjustment factor

to project CCRs” in addition to inflating charges. 71 Fed. Reg. at 48,149. “The commenters

believed that the use of more than one indicator [would] make the threshold calculation more

reliable and accurate.” Id. Thus, these commenters “calculated a cost inflation factor of 5.69

percent by determining the 2002-2004 aggregate annual rate of increase in cost per discharge.”

Id. Another commenter suggested using a “cost inflation factor using the market basket when

projecting CCRs.” Id. at 48,150.

After taking these comments into consideration, HHS agreed in the final rule that “a

refinement to the proposed methodology to account for the rate of change in the relationship

between costs and charges would likely increase the precision of [its] model . . . .” Id. Thus, it

implemented the same fixed loss threshold calculation methodology it had proposed in its notice,

14 except that it used more recent data for the charge inflation factor and that, as commenters

suggested, it “appl[ied] an adjustment factor to the CCRs to account for cost and charge

inflation.” Id. Rather than adopting the precise method proposed by any commenter, HHS

“worked with [its] actual office in deriving” the cost adjustment factor methodology the Court

described above—it “calculated . . . the average increase (over all relevant hospitals) in hospital

operating costs; and it divided that figure by the increase in the ‘market basket’ . . . for the same

time period. . . . HHS then averaged that number over three prior years.” Pls.’ Summ. J. Mem. at

16 (citing 71 Fed. Reg. at 48,150). HHS explained that it believed this calculation was “more

accurate and stable than the commenters’ methodology because it takes into account the costs per

discharge and the market basket percentage increase when determining a cost adjustment factor.”

71 Fed. Reg. at 48,151. Using this method, the final FY 2007 rule adopted a fixed loss threshold

of $24,475, “$1,055 lower than the $25,530 threshold from the proposed rule.” Id.

The final rule’s fixed loss threshold of $24,475, calculated using a CCR adjustment

factor, was a logical outgrowth of the proposed rule’s suggestion of a fixed loss threshold of

$25,530, calculated using CCRs from the most recent available previous year’s data. A final rule

is a logical outgrowth of a proposed rule “[i]f interested parties ‘should have anticipated’ that the

change was possible, and thus reasonably should have filed their comments on the subject during

the notice-and-comment period . . . .” Am. Coke & Coal Chems. Inst. v. EPA, 452 F.3d 930,

938–39 (D.C. Cir. 2006) (cleaned up). “It generally is not a violation of notice and comment

requirements to amend a proposed rule in response to a comment.” Solite Corp. v. EPA, 952

F.2d 473, 496 (D.C. Cir. 1991); see Select Specialty Hosp.-Akron, LLC v. Sebelius, 820 F. Supp.

2d 13, 23 (D.D.C. 2011) (“There is no requirement that an agency ‘select a final rule from

among the precise proposals under consideration during the comment period.’” (quoting Sierra

15 Club v. Costle, 657 F.2d 298, 352 (D.C. Cir. 1981))). Thus, the agency need not “assiduously lay

out every detail of a proposed rule for comment.” Horsehead Res. Dev. Co. v. Browner, 16 F.3d

1246, 1268 (D.C. Cir. 1994). Indeed, “[t]he whole rationale of notice and comment rests on the

expectation that the final rules will be somewhat different—and improved—from the rules

originally proposed by the agency.” City of Stoughton v. EPA, 858 F.2d 747, 753 (D.C. Cir.

1988). The D.C. Circuit’s notice-and-comment cases “focus . . . primarily on whether the final

rule changes critically from the proposed rule . . . . The question is typically whether the

agency’s final rule so departs from its proposed rule as to constitute more surprise than notice.”

Air Transp. Ass’n of Am. v. F.A.A., 169 F.3d 1, 7 (D.C. Cir. 1999) (emphasis added); see Long

Island Care at Home, Ltd., 551 U.S. at 174 (noting that the object of the logical outgrowth

doctrine “is one of fair notice”).

In the notice of proposed rulemaking for FY 2007, HHS opened for comment its proposal

to use a fixed loss threshold of $24,475 as a means of hitting its 5.1 percent target; it also

proposed for comment its method of achieving that target, which included calculating the year-

on-year average annualized rate of change for charges and using the cost-to-charge ratios from

the most recent available historical data. 71 Fed. Reg. at 24,149–50. It explained, in part, that

this method sought to account for the trend of declining cost-to-charge ratios. Id. at 24,150 (“We

note that the case-weighted national average cost-to-charge ratio declined by approximately 1

percent from the March 2005 to the December 2005 update of the Provider-Specific File.

Hospital charges continue to increase at a steady rate of growth between 7 and 8 percent over

each of the last 2 years, resulting in a decline to the cost-to-charge ratios that are used to compute

the outlier threshold. Using lower cost-to-charge ratios from the December 2005 Provider-

Specific File, in combination with the FY 2005 MedPAR claims and inflated charges,

16 contributes to a higher proposed outlier threshold for FY 2007 compared to FY 2006.”).

Interested hospitals should have anticipated that HHS’s proposed threshold and its method of

calculating it might change in some way, including via refinement of the CCR-projection

element of the calculation. See Long Island Care at Home, Ltd., 551 U.S. at 175 (“Since the

proposed rule was simply a proposal, its presence meant that the Department was considering the

matter.”). Any hospital that wished to suggest that the threshold was too high to hit the target, or

to suggest any adjustment to the projection methodology, was on notice of its opportunity to do

so. And that is exactly what hospitals did. HHS’s final methodology built upon these

comments. 71 Fed. Reg. at 48,149–50 (noting in the final rule commenters’ concern[s] that the

proposed threshold was too high and their suggestions for “further refining the outlier

methodology” and agreeing that it would “increase the precision of [HHS’s] model” to “account

for the rate of change in the relationship in costs and charges”—i.e. the precise nature of the

trend of declining CCRs—by “applying an adjustment factor to the CCRs”). “Commenters

clearly understood that” how best to project CCRs to account for the trend of decline was “under

consideration, as the agency received comments [on the method] from several sources.” See

Appalachian Power Co. v. EPA, 135 F.3d 791, 816 (D.C. Cir. 1998).

This case is similar to Solite Corp. v. EPA, 952 F.2d at 485 (per curiam), which

concerned an EPA rulemaking geared at determining whether certain types of industrial wastes

were subject to a particular regulatory category. A waste qualified for the category if it was both

“high volume” and “low hazard.” See id. at 480. In the notice of proposed rulemaking, the EPA

relied on a combination of plant-specific and industry wide data to propose that a solid waste

would qualify as “high volume” if its annual output crossed the threshold of 50,000 metric tons

per year and a liquid waste would qualify if its annual output crossed the threshold of 1.5 million

17 metric tons per year. Id. But during the comment period, “more recent and complete” data

became available that included “detailed information about volumes and specific types of wastes

generated” at relevant facilities. Id. at 481 (citation omitted). In response, the EPA “revised” its

analysis, and the final rule proposed an annual volume threshold for qualification of 45,000

metric tons for solid wastes and one million tons for liquid wastes. Id. (citation omitted). EPA’s

analysis had changed between the proposed and final rule, but its methodology “[had] not

change[d] significantly.” Id. at 485. The new data had allowed EPA to employ “a more precise

quantitative measure based on more complete information” and allowed EPA to adjust in

response to comments critical of the data used in the proposed rule. Id. The D.C. Circuit upheld

the EPA’s rule against a notice-and-comment challenge. Id.

Similarly, in the instant case, HHS proposed a figure (the fixed loss threshold for FY

2007) and explained its methodology for projecting it, including its effort to account for the trend

of decreasing CCRs. Then, in response to analysis received during the comment period, it

adjusted one piece of its overall method. HHS undoubtedly changed its methodology—it did not

“remain constant”—but it did so in a discrete way that “confirmed,” and built upon, its

announced plan to account for declining CCRs. Id. at 485. Where, as here, “an agency’s

analytic task begins rather than ends with a set of forecasts, sound practice would seem to dictate

disclosure of those forecasts so that interested parties can comment upon the conclusions

properly to be drawn from them.” Air Transp. Ass’n of Am., 169 F.3d at 8 (quoting Indep. U.S.

Tanker Owners Comm. v. Lewis, 690 F.2d 908, 926 (D.C. Cir. 1982)).6 HHS did so here, and

6 Air Transport Association of America involved an F.A.A.-specific procedure that is “similar to the notice and comment procedure for informal rulemaking under the Administrative Procedure Act,” so the D.C. Circuit applied logical outgrowth principles drawn from APA cases. 169 F.3d at 6–7.

18 adjusted its forecast methodology in a discrete way that was in line with, though not identical to,

suggestions it received from commenters. Compare Am. Coke & Coal Chems. Inst., 452 F.3d at

940 (“Given that the commenters ‘had a fair opportunity to present their views on how the

industry ought to be subcategorized’ and that the choice to merge the subcategories was a

foreseeable result of the EPA’s solicitation of comment on rationality of the subcategory scheme,

there was no failure of notice or opportunity to comment even though parties may not have been

able to predict how the EPA would choose to act upon their comments.” (quoting BASF

Wyandotte Corp. v. Costle, 598 F.2d 637, 642–46 (1st Cir. 1979))), with CSX Transp., Inc. v.

Surface Transp. Bd., 584 F.3d 1076, 1081–82 (D.C. Cir. 2009) (holding that a final rule

permitting ratemaking parties to rely on four years of comparison data for a certain benchmark

was not a logical outgrowth of a proposal to allow for one year of comparison data, because even

though the rule was not “a complete turnaround from the NPRM,” there was “no way

commenters . . . could have anticipated” that this “particular aspect[] of [the] . . . proposal [was]

open for consideration (citation omitted)).

It is worth pausing further over this point: Plaintiffs are correct that it may not have been

possible for hospitals to predict precisely how HHS would respond to its solicitation for

comments on the CCR projection piece of its methodology for setting the FY 2007 outlier

threshold. They may not have foreseen, for example that HHS would take the three-year average

of the increase in hospital operating costs compared to the market basket increase, rather than, as

some commenters suggested, simply deriving the cost inflation factor from “the annual rate of

increase in the cost per discharge” in recent years. See 71 Fed. Reg. at 48,150. But case law

suggests a distinction between major shifts in direction or policy between proposal and rule,

which are not logical outgrowths, and mere refinements or adjustments consistent with the

19 direction of a proposed policy, which are. CSX Transp., Inc, 584 F.3d at 1081 (noting that “D.C.

Circuit cases finding that a rule was not a logical outgrowth have often involved situations where

the proposed rule gave no indication that the agency was considering a different approach, and

the final rule revealed that the agency had completely changed its position,” and citing cases in

which the final rule “adopt[ed] a maximum velocity cap where a minimum was proposed” and

adopted a “completely different reading of a set of regulatory standards” from the reading

proposed (emphasis added)); Air Transp. Ass’n of Am. v. C.A.B., 732 F.2d 219, 224 (D.C. Cir.

1984) (final rule was a logical outgrowth because the “critical elements of the proposal did not

change”). Thus, in contrast to the D.C. Circuit’s conclusion in Solite Corp. that the change in the

high volume threshold did not violate notice-and-comment requirements because it was merely

the result of EPA’s use of better data “to check or confirm prior assessments,” 952 F.2d at 485,

the same panel held in a later portion of the opinion that the EPA’s final-rule conclusion that a

particular waste would be excluded from the regulatory category at issue was not a logical

outgrowth of its proposed rule, which had listed that waste as a candidate for inclusion in the

regulatory category. The proposed rule had not listed the waste on a separate list of proposed

wastes to be excluded, and did not explain the methodology EPA eventually used to exclude the

waste. Id. at 498–500; compare Husqvarna AB v. EPA, 254 F.3d 195, 199, 203 (D.C. Cir. 2001)

(final rule’s imposition of a four-year implementation period to meet new emission standards

was a logical outgrowth of proposed rule’s five-year period, where EPA decided during

comment period that “‘rapid technological advances’ . . . warranted a more expeditious

implementation”), and Select Specialty Hosp.-Akron, LLC, 820 F. Supp. 2d at 23–24 (rule

imposing discharge limits on certain hospitals-within-hospitals in the first year of a transition

period for implementing a new policy limiting hospitals-within-hospitals’ admissions from their

20 host hospitals to 25 percent of their discharges was a logical outgrowth of proposals that did not

mention the first-year requirements, because these were “merely an implementing mechanism for

the 25 percent rule,” which had been “the subject of extensive public comment”), with Allina,

746 F.3d at 1109 (final rule was not a logical outgrowth because it represented “a volte-face”

from the proposal with “enormous financial implications”), and Air Transp. Ass'n of Am., 169

F.3d at 7–8 (final order partially approving airport’s application to impose a passenger facility

charge did not comply with notice-and-comment procedures where the approval was based on a

projection of passenger increase not included in the initial proposal that was “an order of

magnitude greater” than the increase the initial proposal relied upon and was “a measure of

something completely different”).

HHS’s adjustment of its method of determining a single component of its overall project

of estimating the outlier threshold that would result in outlier payments equal to 5.1 percent of

total payments for the final year, see Billings Clinic, 901 F.3d at 307–09, which adjustment

contributed in part to the final rule’s downward revision of the proposed outlier threshold by

$1,055, fits more neatly into the refinement/adjustment category. See Air Transp. Ass’n of Am.,

732 F.2d at 225 n.12 (concluding that a final fee schedule was a logical outgrowth of a proposal

in part because the changes in fees reflected on the final schedule could not “be considered

major”). Owner-Operator Indep. Drivers Ass’n v. Fed. Motor Carrier Safety Admin., 494 F.3d

188, 202 (D.C. Cir. 2007) at first glance seems to undercut this conclusion, but upon closer

inspection confirms it. There, the D.C. Circuit held that the Federal Motor Carrier Safety

Administration (“FMCSA”) violated the notice-and-comment requirement when it promulgated

a final rule governing rest requirements for truck drivers. Id. The final rule employed a model to

estimate truck drivers’ fatigue by using “time-on-task multipliers,” which sought to account for

21 the fact that time spent driving is more fatiguing than time spent resting. Id. at 200–01. The

flaw: though the FMCSA’s model was an “update” to a safety model it had used in a previous

rule and disclosed in the proposal, the “nature of the update” was to use an “entirely new”

methodology, namely the use of time-on-task multipliers, which were “an integral part” of the

model. Id. at 201.

Here, HHS similarly updated the methodology of a disclosed model between the

proposed and final rule, but HHS’s comment-prompted refinement of a discrete (though

important) factor in a multi-part calculation renders “the nature of the update,” id., different in

kind than FMCSA’s wholesale methodological replacement in Owner-Operator Independent

Drivers Association. For one thing, this portion of the Owner-Operator Independent Drivers

Association opinion did not analyze whether the final rule was a logical outgrowth of the

proposed rule; it instead engaged in the related, but distinct, inquiry of whether the methodology

was “critical factual material” that the agency had to disclose for comment. Id. at 199. More

fundamentally, the D.C. Circuit’s key conclusion—that “[a]lthough interested parties may have

known that FMCSA would incorporate time-on-task effects into its crash-risk model, they had no

way of knowing that the agency would calculate the impact of time on task in the way that it

did”—turned on the conclusion that there was “no way” for commenters “to foresee” four

specific elements of the agency’s methodology, three of which contradicted the approach taken

in supporting studies in the administrative record. Id. at 202. It was only “in light of these

undisclosed elements” that the court concluded that the model had not been “made public in the

proceeding and exposed to refutation.” Id. “Moreover,” the court reasoned, “the addition of the

time-on-task element to the model was not a minor modification used to check or confirm prior

analyses: it constituted the agency’s response to an important defect in its previous methodology

22 identified” in a previous D.C. Circuit decision. Id. at 201. As discussed, commenters foresaw

the possibility that HHS would apply an adjustment factor to CCRs in the FY 2007 final rule,

and though they did not precisely predict HHS’s exact final method for deriving the adjustment

factor, the final method was in line with commenter proposals. Though it is a close question, the

FY 2007 final rule’s refinement of a single element of its method of projecting the fixed loss

threshold was more like a “minor modification used to check or confirm prior analyses,” than the

introduction of an “entirely new” methodology held impermissible in Owner-Operator

Independent Drivers Association, id.

The bottom line is that the $1,055 difference between the proposed and final fixed loss

thresholds, a result at least in part of HHS’s change to its method of projecting CCRs, was

“reasonably foreseeable,” Long Island Care at Home, Ltd, 551 U.S. at 175 , so it did not require

the opening of a new round of notice and comment. See Weyerhaeuser Co. v. Costle, 590 F.2d

1011, 1031 (D.C. Cir. 1978) (an agency “need not subject every incremental change in its

conclusions after each round of notice and comment to further public scrutiny before final

action”).

2. HHS was not required to disclose Attachment A during the FY 2007 Rulemaking.

The Court’s conclusion that the FY 2007 final rule was a logical outgrowth of the

proposed rule dictates the further conclusion that Plaintiffs’ Portland Cement argument against

the FY 2007 rule must fail as well. The D.C. Circuit’s Portland Cement or “critical-material”

doctrine, Post Acute Med. at Hammond, LLC v. Azar, 311 F. Supp. 3d 176, 184–85 (D.D.C.

2018), provides that “[u]nder APA notice and comment requirements, among the information

that must be revealed for public evaluation are the technical studies and data upon which the

agency relies in its rulemaking.” Banner Health, 867 F.3d at 1336 (citation omitted); but see Am.

Radio Relay League, Inc., 524 F.3d at 245–47 (Kavanaugh, J. concurring in part, concurring in

23 the judgment in part, and dissenting in part) (concluding that the Portland Cement doctrine is

inconsistent with the Supreme Court’s holding in Vt. Yankee Nuclear Power Corp. v. Natural

Res. Def. Council, Inc., 435 U.S. 519, 524 (1978) that courts may not impose upon agencies

procedural requirements beyond those found in the APA’s text). Plaintiffs invoke this doctrine

to argue that HHS should have provided the opportunity to comment on “Attachment A,” a

document HHS eventually certified for inclusion in the administrative record in this action for

the final FY 2007 rule. AR (FY 2007) at 14403. According to Plaintiffs’ description of the

document, Attachment A was important to HHS’s decision in the final FY 2007 rule to average

hospital costs compared to general inflation over three years (rather than some other period),

because it shows that HHS ran calculations using different averaging periods:

Attachment A reveals that HHS carried out similar calculations using those other alternative models. In the first table in the document, the column labeled “Oper MB Increase” is evidently the “market basket” (the “MB”) for a given year. AR 14403. The column labeled “Operating Cost per Discharge Increase” is presumably the average hospital operating cost increase over the same year. “Cost/MB” is the number produced by dividing them . . . . “Mean Ratio to 2004” is . . . the average of the “Cost/MB” figures for the number of years specified in the “Num Years in Mean” column. The final column is the result of multiplying the “Mean Ratio to 2004” by the “Oper MB Increase” figure for 2005.

Pls.’ Summ. J. Mem. at 17.7 Thus, Plaintiffs appear to acknowledge (and the document

admits of no other conclusion) that Attachment A is relevant only to calculation of the

7 The Court reproduces the described portion here, AR (FY 2007) 14403:

24 adjustment factor to be applied to CCRs (specifically the selection of a three-year average)—a

choice the Court has already concluded was a logical outgrowth of the proposed rule, and

therefore a choice on which HHS was not required to allow further comment. If HHS was not

required to submit its adjustment-factor method for comment, it follows that it was not required

to submit documents underlying that method for comment. See Air Transp. Ass’n of Am., 732

F.2d at 224 (rejecting APA challenge to agency reliance on internal staff studies not publicly

available during the comment period because the “critical elements of the proposal did not

change, and the final rule was a ‘logical outgrowth’ of the proposed rule”); cf. Bldg. Indus. Ass’n

of Superior Cal. v. Norton, 247 F.3d 1241, 1246 (D.C. Cir. 2001) (“[A] final rule that is a logical

outgrowth of the proposal does not require an additional round of notice and comment even if the

final rule relies on data submitted during the comment period.”); City of Stoughton, 858 F.2d at

753(agency was not required to solicit comment on a study that came to its attention only after

the publication of a proposed rule where the study grounded a final-rule conclusion that was a

logical outgrowth of the proposal because it confirmed the proposal’s conclusion); Chamber of

Com. of U.S. v. SEC, 443 F.3d 890, 900, 903–04 (D.C. Cir. 2006) (distinguishing between

25 “entirely new information critical to the agency’s determination” and “supplementary

information,” including data that “clarif[ies], expand[s], or amend[s] other data that has been

offered for comment,” and noting that absent “a showing of prejudice by an interested party,” the

agency need not allow “further opportunity for comment[] provided that the agency’s response

constitutes a ‘logical outgrowth’ of the rule proposed” (citations omitted and cleaned up)). This

result makes sense, as the purpose of the critical material doctrine is to facilitate comment on a

proposed rule. See Banner Health, 867 F.3d at 1336 (the doctrine “allows for useful criticism,

including by enabling commenters to point out where information is erroneous or where the

agency may be drawing improper conclusions” (citations omitted and cleaned up)). By hewing

sufficiently closely to the proposal to render its final rule a logical outgrowth, the agency has

already complied with its obligation to allow a sufficient opportunity for comment.

B. FY 2008–2013 Rules

1. The Court may consider Attachment A and the associated memorandum in order to evaluate Plaintiffs’ critical-material challenge to the FY 2008–2013 rules.

Plaintiffs next expand their critical material argument against the FY 2007 rule to assert

that HHS violated the Portland Cement doctrine when it failed to disclose Attachment A and the

accompanying memorandum for comment during the rulemakings for FYs 2008–2013. Pls.’

Summ. J. Mem. at 20–22. As Plaintiffs acknowledge, Attachment A and the associated

memorandum (entitled “Change to Outlier Methodology”) are not currently part of the

administrative records for these rules, and their argument on this point “depends on the

assumption that Attachment A and the associated memo are in the record for the FY 2008

through 2013 rules.” Id. at 20 n.12. Therefore, before reaching the merits of this argument, the

Court must first consider, in part, Plaintiffs’ motion to complete the administrative records. In

this motion, Plaintiffs argue that Attachment A and the Change to Outlier Methodology

26 memorandum should be part of the FY 2008–2013 administrative records for all purposes

because HHS either actually considered these documents during these rulemakings or

negligently failed to do so. Pls.’ Mot. Complete Admin. R. at 5–12. In the alternative, they

argue that the Court should at least consider these documents for the limited purpose of

determining whether HHS violated Portland Cement by failing to disclose them during the FY

2008–2013 rulemakings. Id. at 13–16. The Court concludes that it may consider the documents

at least in order to evaluate the procedural Portland Cement claim and accordingly grants the

motion to complete in part.8

8 The parties have vigorously briefed whether Plaintiffs’ motion to complete the administrative record by adding Attachment A and the associated memorandum is timely, but, as the Court sees it, the issue is relatively straightforward. On March 31, 2020, the Court ordered the Secretary to produce the memorandum to Plaintiffs. Mot. Dismiss Op. at 33 (“The Court will direct the Secretary to disclose to Plaintiffs the full document with which “Attachment A” is associated.”). It further held that “[o]nce such document is disclosed, Plaintiffs may, if appropriate, renew their argument that this document should be part of the administrative record.” Id. Thus, the Court contemplated a departure from its previous scheduling order, which set a deadline for supplementation motions of June 5, 2019. ECF No. 137. To be sure, the Court said this in the context of evaluating Plaintiffs’ assertion that the memorandum was before the agency during the FY 2007 rulemaking, and therefore arguably did not refer to a motion to complete the administrative records for FYs 2008–2013. But the Court said Plaintiffs could renew their “argument,” a term broader than the 2007-specific motion, and, given that the opinion dealt with challenges to a range of different years’ rules, the use of the word “record” is fairly read to refer to the entire record before the Court, even though that record is technically comprised of individual records for various rulemakings. Due to intervening litigation about the scope of the required disclosure, the Secretary did not produce the memorandum until April 13, 2021. Though Plaintiffs may have been able to surmise it on their own previously, this production provided—for the first time— record evidence that Attachment A was produced by the actuary’s office in order to develop the cost inflation adjustment factor method disclosed in the final FY 2007 rule. Shortly thereafter, in a joint status report, Plaintiffs informed the Court that they might “need to file a further motion regarding the content of the administrative records . . . .” ECF No. 179 at 2. About three months after the production, on July 16, 2021, Plaintiffs moved to add Attachment A and the memorandum to the FY 2008–2013 administrative records on the theory that the FY 2007 rule relied on these actuary documents, and the subsequent rules referenced the FY 2007 rule’s actuarial analysis as the basis for their own methodologies. Moreover, even though this motion to complete came the same day as Plaintiffs’ motion for summary judgment, the Secretary had two months to account for the filing before submitting

27 The “record rule” generally prohibits the consideration of evidence outside the

administrative record in APA actions: “[I]t is black-letter administrative law that in an APA case,

a reviewing court should have before it neither more nor less information than did the agency

when it made its decision.” Hill Dermaceuticals, Inc. v. FDA, 709 F.3d 44, 47 (D.C. Cir. 2013)

(cleaned up). However, a court may make an exception when “the procedural validity of the

agency’s action remains in serious question.” CTS Corp. v. EPA, 759 F.3d 52, 64 (D.C. Cir.

2014) (citation omitted); see United Student Aid Funds, Inc. v. Devos, 237 F. Supp. 3d 1, 4

(D.D.C. 2017) (describing the exception for cases in which “the district court cannot determine

from the administrative record whether the agency complied with its procedural obligations”

(first citing Esch v. Yeutter, 876 F.2d 976, 991 (D.C. Cir. 1989), and then citing CTS Corp, 759

F.3d at 64)). Still, this exception applies “at most” to challenges to “gross procedural

deficiencies” (citation and emphasis omitted). CTS Corp., 759 F.3d at 65.

If Plaintiffs are right that Attachment A and the Change to Outlier Methodology

memorandum were critical material to the FY 2008–2013 rulemakings and HHS withheld them

from commenter scrutiny, then HHS’s notice-and-comment procedures were grossly deficient.

See Portland Cement, 486 F.2d at 392 (describing the failure to disclose critical material as a

his own summary judgment motion in September. Therefore, unlike in Banner Health v. Burwell, 126 F. Supp. 3d 28, 60 (D.D.C. 2015), aff’d in relevant part, Banner Health, 867 F.3d at 1336, in which the court considered it “tardy” to file a motion for consideration of extra-record material on the same day that both plaintiffs’ and defendant’s motions for summary judgment were due, the Secretary has not been prejudiced. The D.C. Circuit held that this conclusion was within the scope of the district court’s discretion in part because the “delay denied HHS the opportunity to treat the comment as part of the administrative record in preparing its motion for summary judgment.” Banner Health, 867 F.3d at 1336. That is not the case here. “Given the general judicial preference for resolving motions on their merits rather than dismissing them on technicalities,” Niedermeier v. Off. of Baucus, 153 F. Supp. 2d 23, 27 (D.D.C. 2001), under these circumstances, the Court finds that the motion to complete the 2008–2013 rules is timely and will consider its merit.

28 “critical defect”); Solite Corp., 952 F.3d at 484 (describing such a failure a “serious procedural

error” (citation omitted)); Am. Radio Relay League, Inc., 524 F.3d at 237 (“It would appear to be

a fairly obvious proposition that studies upon which an agency relies in promulgating a rule must

be made available during the rulemaking in order to afford interested persons meaningful notice

and an opportunity for comment. It is not consonant with the purpose of a rule-making

proceeding to promulgate rules on the basis of inadequate data, or on data that, [to a] critical

degree, is known only to the agency.” (cleaned up)); Ctr. for Biological Diversity v. U.S. Army

Corps of Eng’rs., No. 20-CV-103, 2020 WL 5642287, at *14–15 (D.D.C. Sept. 22, 2020)

(granting plaintiffs permission to rely on extra-record evidence in order to argue that an agency

failed to comply with applicable notice-and-comment procedures). The Secretary objects on the

ground that CTS Corp. “held that an alleged violation of the [Portland Cement] doctrine did not

justify consideration of extra-record evidence.” Def.’s Admin. R. Opp’n at 13. But that is a

misstatement of the case’s holding—the plaintiff attempted to introduce record evidence not to

show a procedural notice-and-comment defect, but rather to challenge the substance of the

agency’s decision. CTS Corp., 759 F.3d at 63–65 (characterizing the plaintiff’s attempt in a

footnote to frame its argument as procedural as “conclusory,” noting that plaintiff did not seek a

remedy appropriate for a procedural violation, and observing that the plaintiff “could have

pursued a procedural challenge arguing that the EPA’s failure to include [certain] data in the

record at the promulgation stage required that it be afforded an additional opportunity to

comment on the data” but that it did not do so (emphasis added)). Moreover, unlike in this case,

the CTS Corp. plaintiff asked the Court to rely on “its own expert’s newly created analysis”

responding to data in the final administrative record, not to consider data that was within the

agency’s files during the rulemaking and allegedly should have been disclosed for comment. Id.

29 at 65. The Court will consider Attachment A and the Change to Outlier Methodology

memorandum in evaluating Plaintiffs’ critical material challenge to the FY 2008–2013 rules.

2. HHS was not required to disclose Attachment A and the associated memorandum during the FY 2008–2013 rulemakings.

Plaintiffs’ critical material challenge is that together, the memorandum and Attachment A

were critical factual material for the FY 2008–2013 rules because they represented HHS’s

“analysis of alternative methodologies”; that is, they showed HHS’s exploration of the effect of

using averaging periods other than three years on the CCR adjustment factor. Pls.’ Summ. J.

Mem. at 20–22. According to Plaintiffs, Attachment A’s revelation that different averaging

periods resulted in widely varying adjustment factors belies HHS’s assertions in the FY 2008–

2013 rules that its adjustment factor method was “accurate and stable.” Id. at 23–24 (citation

omitted).

Once again, under Portland Cement, the agency must “identify and make available

technical studies and data that it has employed in reaching the decisions to propose particular

rules. An agency commits serious procedural error when it fails to reveal portions of the

technical basis for a proposed rule in time to allow for meaningful commentary.” Owner-

Operator Indep. Drivers Ass’n, Inc., 494 F.3d at 199 (cleaned up). This disclosure rule does not

extend to all data in an agency’s files; rather, it applies to “the ‘most critical factual material’

used by the agency.” See Chamber of Com. of U.S., 443 F.3d at 900 (citation omitted). Thus,

when “staff reports” are critical to the agency’s rule, the agency must submit them for comment.

See Am. Radio Relay League, Inc., 524 F.3d at 236 (citation and emphasis omitted). Assuming

without deciding that HHS considered Attachment A and the Change to Outlier Methodology

memorandum during the FY 2008–2013 rulemakings, as opposed to only during the FY 2007

rulemaking—something the Secretary denies, Def.’s Summ. J. Mem. at 21–22—the Court

30 concludes that neither the analysis of alternative averaging periods found in Attachment A nor

the background material found in the Change to Outlier Methodology memorandum were critical

material the agency was required to publish during the FY 2008–2013 rulemakings.

In the rulemakings for FY 2008–13, HHS proposed to use the same fixed loss threshold

projection methodology, including the same CCR adjustment-factor method, that it had used in

the FY 2007 final rule:

As discussed in the FY 2007 final rule (71 FR 48150), we worked with the Actuary to derive the methodology described below to develop the CCR adjustment factor. For FY 2008, we are proposing to use the same methodology by using the operating cost per discharge increase in combination with the final updated market basket increase determined by Global Insight, Inc., as well as the charge inflation factor described above to estimate the adjustment to the CCRs. By using the market basket rate-of-increase and the increase in the average cost per discharge from hospital cost reports, we are using two different measures of cost inflation. For FY 2008, we determined the adjustment by taking the percentage increase in the operating costs per discharge from FY 2004 to FY 2005 (1.0529) from the cost report and dividing it by the final market basket increase from FY 2005 (1.043). We repeated this calculation for 2 prior years to determine the 3-year average of the rate of adjusted change in costs between the market basket rate-of increase and the increase in cost per case from the cost report (FY 2002 to FY 2003 percentage increase of operating costs per discharge of 1.0721 divided by FY 2003 final market basket increase of 1.041, FY 2003 to FY 2004 percentage increase of operating costs per discharge of 1.0624 divided by FY 2004 final market basket increase of 1.04). For FY 2008, we averaged the differentials calculated for FY 2003, FY 2004, and FY 2005 which resulted in a mean ratio of 1.0203. We multiplied the 3-year average of 1.0203 by the 2006 market basket percentage increase of 1.0420, which resulted in an operating cost inflation factor of 6.32 percent or 1.0632. We then divided the operating cost inflation factor by the 1-year average change in charges (1.0726) and applied an adjustment factor of 0.9912 to the operating CCRs from the Provider-Specific File.

See 72 Fed. Reg. 24,680, 24,837 (May 3, 2007), AR at 00159 (FY 2008 Proposed Rule). It then

adopted this methodology in each final rule for FY 2008–13. See 72 Fed. Reg. 47,130, 47,419

(Aug. 22, 2022); Pls.’ Summ. J. Mem. at 20 n.13 (collecting citations for each year’s final rule).

Thus, each proposal referenced the 2007 rulemaking, and, according to Plaintiffs, incorporated

HHS’s analysis of Attachment A and the covering memorandum during the 2007 rulemaking.

31 During the comment period for the FY 2008 rule, one commenter suggested that HHS’s

method of deriving the CCR adjustment factor was “unnecessarily complicated and [did] not

lead to a more accurate” result; the commenter instead suggested “a methodology that uses

recent historical industry wide average rate of change, similar to the methodology used to

develop the charge inflation factor.” 72 Fed. Reg. at 47,418. The commenter does not appear to

have specifically objected to the selection of a three-year averaging period, except insofar as this

choice was one element of the overall adjustment-factor calculation method the commenter

criticized as too complex. See id.; AR at 728. HHS rejected this complaint on the ground that

incorporating two sources of inflation (cost per discharge and market basket increase) would

improve accuracy and stability:

[W]e believe [our] calculation of an adjustment to the CCRs is more accurate and stable than the commenter’s methodology because it takes into account the costs per discharge and the market basket percentage increase when determining a cost adjustment factor. There are times where the market basket and the cost per discharge will be constant, while other times these values will differ from each other, depending on the fiscal year. Therefore as mentioned above, using the market basket in conjunction with the cost per discharge uses two sources that measure potential cost inflation and ensures a more accurate and stable cost adjustment factor.

72 Fed. Reg. 47,130, 47,418 (Aug. 22, 2007). The records for the FY 2009–2012 rules include

very similar comments and responses. Pls.’ Summ. J. Mem. at 13 (citing 73 Fed. Reg. 48,434

(Aug. 19, 2008) 48,763, AR 4928 (FY 2009 rule); 74 Fed. Reg. 43,754, 44,008 (Aug. 27, 2009),

AR 7080 (FY 2010 rule); 75 Fed. Reg. 50,042, 50,427 (Aug. 16, 2010), AR 7815 (FY 2011

rule); 76 Fed. Reg. 51,476, 51,792 (Aug. 18, 2011), AR 11757 (FY 2012 rule)). In 2013,

commenters suggested a wider range of alternatives for adjusting CCRs (none of which

specifically took issue with the three-year averaging period). But HHS stuck with the method it

had been using since 2007, explaining that it needed to study the proposed alternative methods

further. 87 Fed. Reg. 53,258, 53,694 (Aug. 31, 2012).

32 As these proposals and exchanges make clear, HHS disclosed in detail during every

rulemaking each step of its methodology for projecting the fixed-loss threshold, and in turn each

step of its sub-methodology for deriving the CCR adjustment factor, including the selection of a

3-year averaging period. Commenters understood the methodology and took advantage of the

opportunity to suggest alternatives. HHS’s full disclosure of the methodology (and its results)

was sufficient to its critical-material disclosure obligations. Cf. Air Transp. Ass’n of Am., 732

F.2d at 224 (holding that a final fee schedule that turned in part on agency staff studies not

available during the comment period complied with notice-and-requirements because the

proposal “both outlined the method by which [the agency] proposed to calculate the fees and

listed the types of fees it proposed to charge”); Indep. U. S. Tanker Owners Comm., 690 F.2d at

926 (“[W]here an agency’s analytic task begins rather than ends with a set of forecasts, sound

practice would seem to dictate disclosure of those forecasts so that interested parties can

comment upon the conclusions properly to be drawn from them.”); Chamber of Com. of U.S, 443

F.3d at 900 (observing that a second round of notice and comment is not required when an

agency “merely supplements information in the rulemaking record” by “internally generating

information using a methodology disclosed in the rulemaking record”).

It is not as if the document Plaintiffs contend should have been disclosed, Attachment A,

was central to the methodology. It relates at most to a discrete piece of the methodology—

averaging the increase in operating costs compared to the market basket increase over three years

as opposed to some other time period—and even then arguably consists mostly of evaluation of

the results of alternative averaging periods HHS did not ultimately propose or select, rather than

analysis of the three-year choice itself. AR at 14403; see Pls.’ Summ. J. Mem. at 17 (identifying

“[w]hat HHS did not disclose” as the fact that “it had done comparable calculations using

33 averages over two years, four years, five years, etc.”). HHS was not required to disclose analysis

of methodologies it did not ever propose or adopt as a rule.

As for the Change to Outlier Methodology memorandum itself, in the redacted form in

which it appears in the record before the Court, it contains only an anodyne overview of the

adjustment factor methodology and the reasoning behind it. To a significant extent, this

discussion duplicates the description HHS made public in the proposed rules:

We requested OACT [Office of the Actuary] assistance in developing an adjustment to the cost-to-charge ratios. OACT developed a cost inflation factor that could be used in our model. It combines the rate of increase in cost per case with the rate of increase in the IPPS market basket.

If charges are increasing faster than costs, the cost-to-charge ratios will decline, estimates of cost per case and the outlier threshold will be lower than it otherwise would be in the absence of making such an adjustment. The opposite will occur if costs are increasing faster than charges. For FY 2007, the charge inflation factor is slightly higher than the cost inflation factor resulting in a small reduction to the cost-to-charge ratios.

AR (FY 2007) at 14407. These generalities are not the type of analysis the critical-material

doctrine contemplates. In any event, HHS had publicly explained more or less all of the

reasoning reflected in these paragraphs by the time the FY 2008 proposed rule opened for

comment. See 72 Fed. Reg. 24,680, 24,837 (May 3, 2007), AR (FY 2008) at 00159 (explaining

in the FY 2008 proposed rule that “by using the market basket rate-of-increase and the increase

in the average cost per discharge from hospital cost reports, we are using two different measures

of cost inflation”); 71 Fed. Reg. at 48,150 (explaining in the FY 2007 final rule that HHS

decided to use an adjustment factor in order to “account for the rate of change in the relationship

between costs and charges”).

That HHS publicly disclosed a detailed step-by-step methodology, complete with the

figures yielded at each step, distinguishes this case from the key authorities on which Plaintiffs

rely. In Owner-Operator Independent Drivers Association, Inc., for example, the D.C. Circuit

34 vacated parts of a rule regulating truck-driver rest practices because the agency “fail[ed]” to

“disclose the methodology” of a model that was “central to the agency’s justification for the

rule.” 494 F.3d at 199–200. In particular, the agency failed to disclose its use of certain

multipliers that measured driver fatigue based on time spent driving; these were “an integral

part” of the model. Id. at 201. Here, in contrast, HHS disclosed each step of its methodology,

including the three-year averaging period Plaintiffs contest. Indeed, Plaintiffs do not claim that

HHS failed to disclose its methodology, but rather “its analysis of alternative methodologies.”

Pls.’ Summ. J. Mem. at 20 (cleaned up). Yet they do not cite any authority for the proposition

that, in addition to disclosing “portions of the technical basis for a proposed rule,” Owner-

Operator Indep. Drivers Ass’n, Inc., 494 F.3d at 199 (cleaned up and emphasis added), an

agency must also disclose material related to its evaluation of a rule or methodology it ultimately

chose not to pursue and not to propose to the public. Cf. Chamber of Com. of U.S., 443 F.3d at

902 (finding a critical-material violation where undisclosed materials “supplie[d] the basic

assumptions” the agency used to support its cost-benefit analysis); Penobscot Indian Nation v.

U.S. Dep’t of Hous. & Urb. Dev., 539 F. Supp. 2d 40, 49 (D.D.C. 2008) (finding a critical-

material violation where the agency failed to disclose an internal analysis that was the

“centerpiece” for the final rule’s rationale where the proposed rule neither referenced the

analysis nor included “at least a summary of the specific data and methodology on which the

analysis relied”).

Similarly, in Shands Jacksonville Med. Ctr. v. Burwell, 139 F. Supp. 3d 240, 247, 265

(D.D.C. 2015), the court held that HHS failed to meet its procedural obligations when it relied on

an undisclosed actuary analysis to estimate that a change in how patient stays were classified

would increase the number of patients classified as inpatients, and, in turn, to reduce hospital

35 compensation for inpatient services. The court framed “[t]he question” as “whether the public

was aware of the methodology the HHS actuaries used to predict the effects of” the classification

change. Id. at 264. The court rejected the Secretary’s contention that the proposed rule

discussed the methodology, because all the proposed rule said was that “the Secretary analyzed

‘FY 2009 through FY 2011 Medicare claims data for extended hospital outpatient encounters

and shorter stay hospital inpatient encounters.’” Id. at 263 (quoting the proposed rule). This did

not reveal “critical assumptions” reflected in an actuarial memorandum used in reaching the

estimate, including that the actuarial analysis accounted only for claims related to certain types

of procedures. Id. at 262–64. Thus, there was “no reason to believe that commenters had any

idea what the actuaries did” with the underlying data. Id. at 264. Here, HHS told the public

exactly what its actuaries did, right down to the selection of a three-year averaging period. The

selection of a three-year period may have been an assumption, but unlike in Shands, HHS

disclosed the choice and opened it to commenter criticism.

American Radio Relay League, Inc. v. FCC is probably the strongest case for Plaintiffs,

but it, too, is meaningfully distinct. In that case, the D.C. Circuit took issue with the fact that the

agency’s “determination [was] based upon a complex mix of controversial and uncommented

upon data and calculations.” 524 F.3d at 237 (citation omitted and cleaned up). The agency had

disclosed parts of studies that it admitted were “a central source of data for its critical

determinations,” but had redacted other parts of the studies. Id. at 238. These hidden portions

“appear[ed] to contain information in tension with the Commission’s conclusion,” id. (citations

omitted and cleaned up); similarly, Plaintiffs contend that Attachment A’s demonstration that the

adjustment factor “could fluctuate significantly” depending on the averaging period chosen “was

36 contrary to HHS’s assertion that the method was accurate and stable.” Pls.’ Summ. J. Mem. at

22.9

Though the American Radio Relay League panel made some sweeping statements during

the course of its discussion, e.g., 524 F.3d at 238 (“It would appear to be a fairly obvious

proposition that studies upon which an agency relies in promulgating a rule must be made

available during the rulemaking in order to afford interested persons meaningful notice and an

opportunity for comment.”), it went on to emphasize the “narrowness of [its] holding,” which

turned in significant part on the fact that “the Commission ha[d] chosen to rely on” the studies

“and to place them in the rulemaking record” while redacting portions that undermined its

position. Id. at 239. In other words, the FCC could not play “hide and seek” by “cherry-

pick[ing]” only the favorable parts of “a study on which it ha[d] chosen to rely in part.” Id. at

237, 239. This feature was a “critical distinction” in American Radio Relay League, id. at 239,

and nothing like it has occurred here. HHS did not present to the public part of a study in

support of its fixed-loss threshold projection method or its three-year average choice while

carving out Attachment A for exclusion from the study. Indeed, HHS did not support its “more

accurate and stable” assertions by referencing its three-year averaging period at all; read in

context, these turned on the fact that unlike the commenter’s proposed method, HHS’s

adjustment factor took into account “two sources that measure potential cost inflation,” costs per

discharge and the market basket. See, e.g., 72 Fed. Reg. at 47,419 (“[W]e believe [our]

9 As the Court will explain, it is not convinced that Attachment A undercuts HHS’s “more accurate and stable” assertions, which in context did not reference the three-year averaging period but rather compared HHS’s method of accounting for both costs per discharge and the market basket increase to a commenter proposal that would not have taken into account “two sources that measure potential cost inflation.” 72 Fed. Reg. at 47,418.

37 calculation of an adjustment to the CCRs is more accurate and stable than the commenter’s

methodology because it takes into account the costs per discharge and the market basket

percentage increase when determining a cost adjustment factor. There are times where the

market basket and the cost per discharge will be constant, while other times these values will

differ from each other, depending on the fiscal year. Therefore as mentioned above, using the

market basket in conjunction with the cost per discharge uses two sources that measure potential

cost inflation and ensures a more accurate and stable cost adjustment factor.”). Attachment A,

therefore, is not especially relevant to the primary claim Plaintiffs suggest it undermines. See

Pls.’ Summ. J. Mem. at 22.

In sum, in the 2008–2013 rulemakings, HHS disclosed its overall methodology and each

step thereof. The APA did not require it to disclose an analysis of ultimately unchosen

alternatives to a particular step.

3. The FY 2008 final rule’s use of a positive CCR adjustment factor did not necessitate a further round of notice and comment.

Plaintiffs attempt one final challenge to the FY 2008 rule in particular. That year, HHS

proposed to use the adjustment-factor method it had developed in the FY 2007 final rule, which,

when applied to available data from the December 2006 update to the provider-specific file,

produced an adjustment factor of 0.9912. 72 Fed. Reg. at 24,837, AR 159. In other words, the

model forecast that CCRs would be “almost 1% lower in FY 2008” than those reflected in the

latest provider-specific file data. Pls.’ Summ. J. Mem. at 18. In the final FY 2008 rule, HHS

used the very same model, but, as was its general practice, fed into the model more recent data

that had become available during the comment period. See 72 Fed. Reg. at 47,418 (describing

use of the March 2007 update to the provider-specific file and noting different figures for various

inputs into the adjustment factor calculation methodology); see generally Pls.’ Summ. J. Mem. at

38 5 n.6. The result was an adjustment factor of 1.0027—a projection, contrary to the projection in

the proposed rule, that CCRs would increase. 72 Fed. Reg. at 47,418. Plaintiffs claim that the

final rule’s application of a positive adjustment factor was an unfair surprise that required further

opportunity for comment, especially given that HHS had in the proposed rule explained that it

sought to reflect the trend of declining CCRs. See Pls.’ Summ. J. Mem. at 18–20 (citing 72 Fed.

Reg. at 24,837, AR 159).

This challenge fails for the same reason Plaintiffs’ critical-material challenges failed:

HHS disclosed each step of the methodology it would use to set the final CCR adjustment factor.

Regulated entities had ample opportunity to comment on and suggest changes to this

methodology That the methodology’s application to the latest data produced a surprising

outcome did not obligate HHS to open a new round of notice-and-comment proceedings. The

very authorities on which Plaintiffs rely for their challenge demonstrate that HHS met APA

procedural requirements here. Plaintiffs stress that “HHS was ‘required to disclose what the

actuaries did with [the] data”—here HHS did so, and the actuaries’ methods did not change

between the proposal and final rule. Pls.’ Summ J. Mem. at 19 (quoting Shands Jacksonville,

139 F. Supp. 3d at 264). Plaintiffs also rely heavily on a case in which an agency expanded,

without warning, from a methodology that used one year of comparison data to one that used

four years of comparison data. CSX Transp., Inc., 584 F.3d at 1081–82. In this case, HHS did

not change its data parameters (e.g., it used a three-year averaging period the entire time); all that

changed was the data the actuaries operated upon, and, accordingly, the model’s numeric output.

Having disclosed its model, HHS was not required to provide specific notice of the model’s

result when applied to updated data. See Am. Coke & Coal Chems. Inst., 452 F.3d at 939 (“It is

perfectly predictable that new data will come in during the comment period, either submitted by

39 the public with comments or collected by the agency in a continuing effort to give the regulations

a more accurate foundation . . . . If data used and disclosed for the interim regulations presented

the issues for comment, then there is no need to seek new comment even though significant

quantitative differences result.” (quoting BASF Wyandotte Corp. v. Costle, 598 F.2d 637, 644–45

(1st Cir.1979)); cf. Post Acute Med. at Hammond, LLC v. Azar, 311 F. Supp. 3d 176, 183

(D.D.C. 2018) (“The Administrative Procedure Act . . . does not require an agency to advise

regulated entities as to the individualized implications of a proposed rule—particularly here,

where the rule merely continued a longstanding policy with updates reflecting new data.”).

Admittedly, HHS did not explain with great clarity in the FY 2008 final rule which

particular data changed and why. Compare, e.g., 72 Fed. Reg. at 47, 418 (noting that the 2006

market basket percentage increase in the final rule was 1.0430), with, e.g., id. at 24, 836 (noting

that the 2006 market basket percentage increase in the proposed rule was 1.0420). But any

failure in explanation would sound as a substantive defect, not a procedural failure to comply

with notice-and-comment requirements. Plaintiffs have not argued that any failure to explain the

data changes reflected in the FY 2008 final rule was arbitrary and capricious. See Pls.’ Summ. J.

Mem. at 26 (“The Hospitals are not here contending . . . that it was arbitrary for the model to

produce a particular CCR adjustment factor (like the increase forecasted for FY 2008.”)).

Accordingly, the FY 2008 rulemaking was procedurally sound.

V. ANALYSIS: SUBSTANTIVE CHALLENGES

In addition to their notice-and-comment challenges, Plaintiffs claim that HHS acted

arbitrarily and capriciously in various ways when it set the fixed loss thresholds for FYs 2007–

2016. See 5 U.S.C. § 706(2)(A). “A rule is arbitrary and capricious if the agency: (1) ‘has relied

on factors which Congress has not intended it to consider,’ (2) ‘entirely failed to consider an

40 important aspect of the problem,’ (3) ‘offered an explanation for its decision that runs counter to

the evidence before the agency,’ or (4) ‘is so implausible that it could not be ascribed to a

difference in view or the product of agency expertise.’” United States Sugar Corp. v. EPA, 830

F.3d 579, 606 (D.C. Cir. 2016) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm

Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). “Nor will [the Court] uphold agency action if it

fails to consider significant and viable and obvious alternatives.” Dist. Hosp. Partners, L.P. v.

Burwell, 786 F.3d 46, 59 (D.C. Cir. 2015). “[T]he party challenging an agency’s action as

arbitrary and capricious bears the burden of proof.” City of Olmsted Falls v. F.A.A., 292 F.3d

261, 271 (D.C. Cir. 2002) (quoting Lomak Petroleum, Inc. v. FERC, 206 F.3d 1193, 1198 (D.C.

Cir. 2000)) (alteration in original). Though Plaintiffs fail to meet this burden for many of their

challenges, they have demonstrated that the FY 2012 and FY 2013 rules were arbitrary and

capricious in certain respects.

1. HHS’s use of a three-year averaging period when calculating the CCR adjustment factor for FYs 2007–2013 was not arbitrary and capricious.

Plaintiffs first argue that HHS’s decision to average the increase in hospital operating

costs compared to the market basket increase over three years, as opposed to over some other

term of years, was arbitrary and capricious. Plaintiffs say that the possibility of using an

alternative averaging period was a “significant and viable and obvious alternative[],” Pls.’

Summ. J. Mem. at 23 (quoting Dist. Hosp. Partners, L.P, 786 F.3d at 59), and that “the

possibility that a different averaging period might produce a significantly different reduction in

CCRs, compared to the method that HHS did use, [was] surely an important aspect of the

decision-making” that HHS was required to consider, id. The Court disagrees and holds that the

decision to use a three-year averaging period was not significant enough in the context of each

year’s rulemaking to require HHS to consider and explain this particular choice.

41 Plaintiffs have not persuasively argued that whether to average the comparison of

operating cost increases to market basket increases over three years rather than, say, over two or

four years, was a significant aspect of the problem HHS confronted during each rulemaking.

The problem HHS confronted was where to set the fixed loss threshold in order to achieve outlier

payments equal to 5.1 percent of total payments for the coming year. As the Court has

explained, taking the three-year average was but one sub-step in the multi-step process of

projecting the fixed-loss threshold for a given year. Significant problems, by contrast, are those

that are central to the overall objective of the rule. See Util. Solid Waste Activities Grp. v. EPA,

901 F.3d 414, 429–30 (D.C. Cir. 2018) (holding that it was arbitrary and capricious for the EPA

not to consider, when crafting a rule regulating the disposal of toxic wastes produced by

electrical plants, neither “the risks to public health and to the environment before leakage is

detected, nor the harms from continued leakage during the years before leakage is ultimately

halted by retrofit or closure”). Here, whether to account both for operating cost increases and

market basket increases may have been a significant aspect of the problem, but the granular sub-

issue of whether to average that comparison over two, three, or four years was not. See

Michigan v. EPA, 213 F.3d 663, 690–91 (D.C. Cir. 2000) (noting, in the context of rejecting a

challenge to various EPA cost assumptions where the challenger had not explained why the

assumptions were arbitrary and capricious, that courts “generally defer[] to the agency’s

expertise” on “technical details”). This is especially so where plaintiffs have not demonstrated

(or attempted to demonstrate), at least without referencing Attachment A, that an alternative

averaging period would obviously have better helped HHS reach its 5.1 percent target. Pls.’

Summ. J. Mem. at 23 (noting that various alternative periods would have resulted in larger

downward adjustments in CCRs for the FY 2007 and FY 2008 rules without explaining why

42 these adjustments would have been more accurate than the one produced by the three-year period

and conceding that “the point here is not that the Court should determine a two-year or any

particular averaging period would have been superior”).

The Court’s conclusion would hold even were it to consider Attachment A, which shows

that, at least in 2007, HHS was aware that the selection of a particular averaging period could

change the adjustment factor. Plaintiffs rely on Attachment A to demonstrate that “a four-year or

a two-year [averaging period] . . . would, if anything, have produced adjustment factors closer to

the historical trends” and that “different averaging periods could have resulted . . . in lower

thresholds.” Pls.’ Summ. J. Mem. at 22. But matching historical trends is not HHS’s end goal in

setting the fixed-loss threshold; rather, attempting to match historical trends in CCRs is a means

of meeting HHS’s statutory mandate to project payments for the coming year. See, e.g., 71 Fed.

Reg. at 48,151 (explaining the decision to apply an adjustment factor “so that the CCRs we are

using in our simulation more closely reflect the CCRs that will be used in” the coming year); cf.

Banner Health, 867 F.3d at 1355 (“There may well be many non-arbitrary reasons for predicting

that costs and charges in a particular industry will not continue on their current trajectories.”). In

any event, “imperfection alone does not amount to arbitrary decision-making . . . [an] agency

may use a model ‘even when faced with data indicating that it is not a perfect fit.’” Dist. Hosp.

Partners, 786 F.3d at 61–62 (citation omitted). In short, “the Hospitals fail to show why [HHS]

should not receive the deference typically accorded in this context,” Banner Health, 867 F.3d at

1356, and the Court defers to the agency on the technical determination of whether three years

was a an appropriate period over which to compare operating cost and market basket increases.10

10 Plaintiffs also assert that it was arbitrary and capricious for HHS to repeatedly assert that its methodology was “accurate and stable” “when the outcome of the method could be dialed up or down by nearly a factor of ten by changing an arbitrary parameter (the averaging period).

43 Now, we do not know why HHS selected a three-year averaging period; there is no

disputing that HHS never explained this choice. But HHS “was not required to . . . ‘consider all

policy alternatives in reaching its decision.’” Dep’t of Homeland Sec. v. Regents of the Univ. of

Cal., 140 S. Ct. 1891, 1914–15 (2020) (quoting State Farm, 463 U.S. at 51). It follows that HHS

was not required to explain choices that did not rise to the level of significant aspects of the

problem or to explain why it did not select alternatives (e.g., averaging over periods other than

three years) that were not “significant and viable and obvious,” Dist. Hosp. Partners, L.P., 786

F.3d at 59. Cf. Texas Mun. Power Agency v. EPA, 89 F.3d 858, 869–70 (D.C. Cir. 1996) (“And

though the EPA did not explain its precise method for calculating a rate based on a statewide

average that was used in this case until after the close of general proceedings before the agency,

the failure of an agency to identify every detail of a process before it is used does not

automatically require judicial interference in matters that must be thought to lie within the

agency’s expertise.”).11

Pls.’ Opp’n Sec’y’s Cross-Mot. and Reply Supp. Mot. Summ. J. at 28 (“Pls.’ Combination Summ. J. Opp’n & Reply”), ECF No. 193. But as the Court has explained, this takes HHS’s “accurate and stable” assertions out of context. See supra at 37. These did not relate to the choice of averaging period or even assert as a general matter that the model was maximally accurate and stable. Instead, they responded to a particular commenter’s suggestion to rely on historical trends, and noted that using two inflation measures improved accuracy and stability as compared to this suggested method. 72 Fed. Reg. at 47,418. Plaintiffs do not explain how the choice of a three-year averaging period undercuts this specific assertion. 11 Because this claim is the sole remaining one for which consideration of Attachment A and the Change to Outlier methodology memorandum might make any difference, and because the Court would reject this claim whether or not Attachment A and the memorandum are in the administrative records, the Court denies the remaining portion of Plaintiffs’ Motion to Complete Administrative Records—which seeks to have the documents added to the records for all purposes, not just for evaluation of Plaintiffs’ procedural claims—as moot. See Pls.’ Mot. Complete Admin. R. at 5–12; cf. County of Los Angeles, 192 F.3d at 1021–22 (finding it unnecessary to decide whether an affidavit that had “surfaced for the first time during litigation” was a “post-hoc rationalization” that should have been stricken from the record because even were it to consider the affidavit, the Court would still have found against the party offering it).

44 2. HHS’s reasoning for rejecting the use of an “estimate adjustment factor” was sufficient to support the FY 2011 and FY 2014 rules, but was not sufficient to support the FY 2012 and FY 2013 rules.

Plaintiffs find some, though not complete, success with their next set of arguments: that

HHS acted arbitrarily in the FY 2011–2014 rulemakings when it refused to adopt a commenter’s

repeated suggestion to apply an “estimate adjustment factor” that the commenter said would

improve the accuracy of the fixed-loss threshold projection. Pls.’ Summ. J. Mem. at 32–34. The

commenter first suggested the application of an estimate adjustment factor—not to be confused

with the adjustment factor applied to CCRs, which the Court has discussed at length—during the

comment period for the FY 2011 rule. The commenter presented data which it said

demonstrated that HHS had fallen short of its 5.1 percent target for outlier payments each year

since 2004. AR (FY 2011) at 09476–77. The commenter suggested that HHS should recognize

this consistent shortfall and quantify it by averaging the percentage shortfall from the 5.1 percent

target over two or three past years. Id. For example, the commenter calculated that the average

shortfall from the 5.1 percent target was .385 percent over 2008 and 2009. Thus, for 2011, the

commenter recommended modelling as usual, but setting a fixed loss threshold that HHS

estimated would result in outlier payments equal to 5.485 percent of total payments (the 5.1

percent target plus the .385 percent estimate adjustment factor). In other words, HHS’s ultimate

goal would remain to achieve 5.1 percent, and it would use the same model it had used in the

past, but it would attempt to adjust for whatever imperfections in the model had caused it to

undershoot by a .385 percent average over the previous period by directing the model to achieve

an artificially high percentage. See id. The commenter further suggested that if HHS ever faced

the problem of its model consistently overshooting its target, it could apply a negative estimate

adjustment factor using the same method. Id. As the Secretary puts it, the commenter suggested

45 that if HHS declined to change specific asserted defects in the model, it should use a “fudge

factor” designed to correct the model’s end result. See Def.’s Summ. J. Mem. at 28.

HHS declined to adopt this proposal for the FY 2011 final rule, suggesting that it may

have been inconsistent with HHS’s statutory mandate:

[U]nder the statute, outlier payments are intended to approximate the marginal cost of providing care above the outlier fixed-loss cost threshold. Any “estimate adjustment factor” to the outlier threshold or standardized amount in a given year to account for “overpayments” or “underpayments” of outliers in other years would result in us making outlier payments that were not directly related to the cost of furnishing care in extraordinarily costly cases.

75 Fed. Reg. 50,042, 50,429 (Aug. 16, 2010). Moreover, HHS, perhaps understanding the

commenter’s suggestion to relate to revising past outlier payments in addition to projecting

future ones, reiterated its longstanding policy of not making “retroactive adjustments to outlier

payments to ensure that total outlier payments in a past year are equal to 5.1 percent of total

DRG payments.” Id. HHS further asserted that the commenter’s suggestion did “not lend

greater accuracy to [its] estimate of payments that are 5.1 percent of total DRG payments”

because HHS’s model already “factor[ed] in all payments and policies that would affect actual

payments for the fiscal year at hand . . . .” Id.

The commenter tried again during the FY 2012 rulemaking by repeating its estimate-

adjustment-factor suggestion in similar terms. It also attempted to assuage HHS’s statutory

concerns by “stat[ing] that [the proposed estimate adjustment factor method] would fulfill the

statutory requirement in section 1886 (d)(5)(A) of the Act that requires that CMS establish

thresholds such that outlier payments will be projected to achieve at least 5.1 percent of DRG

payments and would more closely achieve a result that is fully consistent with the statute.” 76

46 Fed. Reg. 51,476, 51,793 (Aug. 18, 2011) (describing the comment).12 HHS “thank[ed] the

commenter for further explaining their position on this adjustment” but declined to adopt the

adjustment in two short sentences:

Further analysis by CMS is necessary to determine if the commenter’s approach to applying an “estimate adjustment factor” is appropriate. We will consider the commenter’s suggestion of applying an “estimate adjustment factor” in future rulemaking if, based on our analysis, we determine that application of an “estimate adjustment factor” is appropriate and consistent with the statute.

Id. at 51,794.

Undeterred, the commenter presented the same suggestion again during the FY 2013

rulemaking, complete with data it said showed consistent underpayment. AR (FY 2013) at

13100–01, 13157. But HHS was not swayed, and essentially recycled its 2012 explanation to

again reject the estimate-adjustment-factor proposal in the FY 2013 final rule:

With regard to the comment that CMS implement an “estimate adjustment factor”, as we stated last year, further analysis by CMS is necessary to determine if the commenter’s approach to applying such a factor is appropriate. We will consider the commenter's suggestion to apply an “estimate adjustment factor” (in conjunction with analyzing the alternative methodologies to adjust the CCRs discussed above), for future rulemaking if, based on our analysis, we determine that application of an “estimate adjustment factor” is appropriate and consistent with the statute.

77 Fed. Reg. 53,258, 53,694 (Aug. 31, 2012).

Finally, the commenter tried again in 2014. AR (FY 2014) at 15864–65. The FY 2014

final rule again did not employ an estimate adjustment factor; this time, HHS did not even

12 Both parties say that the FY 2012 comment can be found at pages 11339, 11404–05 of the FY 2012 administrative record, but the Court has been unable to locate these pages in either the original or corrected submissions of the FY 2012 administrative record. However, HHS described the comment in some detail in the FY 2012 final rule, and this description substantially matches the content of the comment submitted for FY 2013. Moreover, the parties do not suggest in their briefing that there are any material differences between the FY 2012 and FY 2013 comments.

47 address the comment. However, HHS made other significant changes to its fixed-loss threshold

projection methodology, which the Court will discuss below. Plaintiffs claim that each of HHS’s

explanations (and the 2014 non-explanation) for rejecting the use of an estimate of adjustment

factor was arbitrary and capricious. Pls.’ Summ. J. Mem. at 32–34. The Court agrees that the

2012 and 2013 explanations, but not the 2011 and 2014 explanations, were insufficient.

FY 2011. To begin with, it is not entirely clear that Plaintiffs are challenging HHS’s FY

2011 explanation for declining to adopt the estimate adjustment factor. They note in their

opening brief that HHS indicated in 2011 that it “thought the method was contrary to the

statute,” and say no more about this before shifting to discussion of the FY 2012 rulemaking in

the very next sentences: “But in the FY 2012 rulemaking, the commenter explained that HHS

had misunderstood the method and it was actually consistent with the statute. HHS accepted that

explanation, thus removing its prior ground for rejecting the proposed method. . . . Why, then,

would HHS not adopt it?” Pls.’ Summ. J. Mem. at 33. Plaintiffs then focus on the assertion that

“[b]y the FY 2012 rulemaking” HHS’s model had missed its target for five years, and complain

that in this context HHS’s vague indications that it needed to address the issue further (and its

failure to say anything at all in the FY 2014 rule) were inadequate—all complaints specific to

HHS’s responses in 2012–2014, but not to relevant to its 2011 response. Id. at 33–34 (emphasis

added). In his combination response brief and memorandum in support of his cross-motion for

summary judgment, the Secretary does not expressly defend HHS’s FY 2011 statutory

explanation—though he does describe it—and instead focuses on arguing that the comment “did

not demonstrate, and did not even purport to demonstrate, that the proposed ‘estimate adjustment

factor’ would operate as a reliable predictor of future differences between projected and actual

payments.” Def.’s Summ. J. Mem. at 29. But then, in their combination response to the

48 Secretary’s cross-motion for summary judgment and reply in support of their own motion for

summary Judgment, Plaintiffs say for the first time that part of HHS’s FY 2011 statutory

reasoning—that the estimate adjustment factor method was inconsistent with the statute because

it “would mean basing the coming year’s forecast on something other than ‘all payments and

policies that would affect actual payments’ for the coming year”—“was arbitrary and

capricious.” Pls.’ Combination Summ. J. Opp’n & Reply at 34 (quoting 75 Fed. Reg. at 50,429).

Use of historical data, Plaintiffs point out in that brief, is core to HHS’s mandate to “‘select

outlier thresholds which, when tested against historical data, will likely produce’” outlier

payments in the targeted range. Id. (quoting County of Los Angeles v. Shalala, 192 F.3d at 1013)

(emphasis added in Plaintiffs’ brief).

Assuming Plaintiffs have properly presented an argument against the FY 2011 response,

the Court rejects it. For one thing, Plaintiffs have not extensively engaged with HHS’s 2011

statutory position; for example, they have not addressed whether the position represented a

reasonable interpretation of an ambiguous statute and is subject to Chevron deference. The

Court will not raise these issues in the first instance. More importantly, Plaintiffs have not

presented any argument that HHS’s other, non-statutory reason for rejecting the estimate

adjustment factor for FY 2011—its assertion that the method would not improve the projection’s

accuracy because HHS’s model already took into account all relevant payments and policies—

was insufficient. 75 Fed. Reg. at 50,429. Plaintiffs’ filings have not convinced the Court that

HHS acted in an arbitrary and capricious manner when it declined to employ an estimate

adjustment factor for FY 2011.13

13 At this point, the Court has rejected on the merits all of Plaintiffs’ challenges to the rules for FYs 2007–2011. Therefore, the Court need not and does not address the Secretary’s alternative argument that even if the challenges to these rules prevail on the merits, certain

49 FY 2012 and FY 2013. The result is different for FYs 2012 and 2013. “An agency is

required to consider responsible alternatives to its chosen policy and to give a reasoned

explanation for its rejection of such alternatives.” Spirit Airlines, Inc. v. United States Dep’t of

Transp., 997 F.3d 1247, 1255 (D.C. Cir. 2021) (quoting Am. Radio Relay League, Inc., 524 F.3d

at 242). In FY 2012, HHS was no longer laboring under its apparent misimpression that the

estimate adjustment factor proposal involved revising past outlier payments. See 76 Fed. Reg. at

51,793–94. HHS did say that it still needed to determine whether the estimate adjustment factor

method was “consistent with the statute,” but it did not say whether this concern rested on either

of the statutory points HHS had raised in 2011—which it did not repeat—or on some other

interpretative concern. Id. at 51,794. More generally, the only information contained in HHS’s

response was that it would consider the suggestion “in future rulemaking if, based on our

analysis, we determine that application of an ‘estimate adjustment factor’ is appropriate and

consistent with the statute.” Id. This was insufficient. See Spirit Airlines Inc., 997 F.3d at 1255

(holding that an agency’s statement that it “plan[ned] to assess” an issue commenters had raised

fell “well short of what [was] needed to demonstrate the agency grappled with an important

aspect of the problem before it or considered another reasonable path forward”). Given the

nature of the comments, HHS was required to say something about why it needed more time to

study the statute than the comment period had allowed. The estimate adjustment factor method

went to the core of the issue before HHS (the accuracy of its projections), was relatively simple

(it allowed HHS to retain the same model it had been using), and was accompanied by

concerning data purporting to show that HHS had consistently missed its targets. See id. at

plaintiffs’ “claims for fiscal years 2007 to 2011 . . . are barred by issue preclusion based on earlier litigation in Banner Health or Billings Clinic.” Def.’s Summ. J. Mem. at 40, 43–45.

50 51,793 (“Based on actual payments determined by the commenter using data analysis, the

commenter asserted that the underpayment has exceeded 0.5 percent in all years except one.”);

see also Thompson v. Clark, 741 F.2d 401, 409 (D.C. Cir. 1984) (an agency’s failure to respond

to a comment is problematic to the extent that “it demonstrates that the agency’s decision was

not ‘based on a consideration of the relevant factors’ (citation omitted)). HHS did not refute or

otherwise engage with this data. The FY 2013 response, which more or less repeated the FY

2012 response, 77 Fed. Reg. at 53,694, rests on even shakier ground: by the time it issued this

response, HHS had had an entire year (at least) to analyze the proposal. If HHS still needed

more time to evaluate, it should at least have explained why. The FY 2012 and 2013 responses

“provide[] no basis upon which [the Court can] conclude that” the decision not to adopt an

estimate adjustment factor for those years “was the product of reasoned decisionmaking.” Cf.

Butte County v. Hogen, 613 F.3d 190, 195 (D.C. Cir. 2010) (citation omitted).

The Secretary’s most promising argument in support of its explanations is that the

comments and their supporting data were not significant enough to merit a response more

detailed than the one HHS gave. The Secretary insists that “[t]he comments . . . did not

demonstrate, and did not even purport to demonstrate, that the proposed ‘estimate adjustment

factor’ would operate as a reliable predictor of future differences between projected and actual

payments.” Def.’s Summ. J. Mem. at 29. But read in context, the comments at the very least

rely on a strongly implied premise that deriving the estimate adjustment factor from past years’

shortfalls would create a factor in line with the model’s past trend of failure, and therefore would

reasonably anticipate the level of failure the model would produce when applied to the coming

year. AR (FY 2013) at 13100–01 (comment submitted for FY 2013); 76 Fed. Reg. at 51,793

(“The commenter stated that this would fulfill the statutory requirement in section 1886

51 (d)(5)(A) of the Act that requires that CMS establish thresholds such that outlier payments will

be projected to achieve at least 5.1 percent of DRG payments and would more closely achieve a

result that is fully consistent with the statute.”).

Similarly, the Secretary is simply wrong that the comments were bereft of “evidence or

analysis supporting the use of an ‘estimate adjustment factor.’” Def.’s Summ. J. Mem. at 29.

The comments presented an analysis and table purporting to show that “actual outlier payments

have averaged more than 0.5 percentage points below the 5.1% target of actual total DRG

payments” in most years since 2003 (FY 2013 comment) or 2004 (FY 2012 comment). AR (FY

2013) at 13100–01, 13,157; 76 Fed. Reg. at 51,793.14 They also provided concrete examples to

demonstrate how the proposed estimate adjustment factors for each year matched recent

underpayments. AR (FY 2013) at 13101 (“For example, the average shortfall from FYs 2009

and 2011 was 0.68 percentage points based on the VHDC studies. The threshold should be

modeled to produce a payment level of 5.78 percent for FY 2013.”); 76 Fed. Reg. at 51,793

(“The commenter provided an example and computed this factor for FY 2009 and FY 2010 by

taking the average variance in the actual payment for FY 2008 and FY 2009 which was 0.491

percent. Based on this factor, CMS would model the threshold to a level of 5.591 percent (5.1

plus .491 percent).”).

Thus, the FY 2012 and 2013 comments were considerably more developed in their data

support and plausibility than comments which courts have held not significant enough to require

an in-depth response. See Thompson, 741 F.2d at 409 (“None of the comments singled out by

appellant as raising substantial issues contained any meaningful analysis or data refuting the

14 The Court does not mean to suggest that HHS was required to accept the commenters’ analysis of this data, but the presentation was plausible enough that HHS should at least have engaged with it.

52 agency’s conclusions.”); Am. Great Lakes Ports Ass’n v. Zukunft, 296 F. Supp. 3d 27, 53–54

(D.D.C. 2017) (holding that the an agency’s failure to respond to a commenter suggestion that it

include a “truing up” mechanism to compensate rate payers who had overpaid due to “disparities

between projections and actual data” did not violate the APA because the suggestion was

presented only obliquely and in one sentence of a forty-page comment) (citation omitted)), aff’d,

962 F.3d 510 (D.C. Cir. 2020). If HHS disagreed with the commenter’s premise and/or the

evidence supporting it, it should have explained why. To the extent that the Secretary now seeks

to raise concerns about the method and data that he did not even hint at in the FY 2012 and 2013

rules—such as his suggestion that the adjustment factor would have “introduc[ed] a new source

of potential distortion,” Def.’s Reply Mem. Supp. Cross Mot. Summ. J. at 19, ECF No. 196—

these are “impermissible post hoc rationalizations and thus are not properly before” the Court,

Dep’t of Homeland Sec., 140 S. Ct .at 1909. “In sum,” the estimate adjustment factor

“alternative was neither frivolous nor out of bounds and the [the agency] therefore had an

obligation to consider it.” Chamber of Com. of U.S. v. SEC, 412 F.3d 133, 145 (D.C. Cir. 2005).

The FY 2012 and 2013 responses do not reflect reasoned consideration of this alternative.

FY 2014. Finally, the Court holds that the FY 2014 rule was not arbitrary and capricious

insofar as it did not adopt or discuss the suggested estimate adjustment factor (which the

commenter renewed during the FY 2014 comment period). At first, HHS’s omission of any

discussion of the estimate-adjustment-factor proposal, which it had for two years promised to

study, seems like a shocking oversight that makes the FY 2014 response even more arbitrary and

capricious than the FY 2012 and FY 2013 responses. But a closer look at the comment in

context reveals that HHS did in effect respond to it.

53 As presented during the FY 2014 rulemaking, the estimate-adjustment-factor suggestion

was just one part of a lengthy comment that suggested several alternative methods for improving

the accuracy of HHS’s fixed loss threshold projections. AR (FY 2014) at 015852–65. The

commenter proposed adoption of an estimate adjustment factor only as a fallback alternative to

several preferred methods. The commenter wrote that it “would prefer that CMS implement the

other improvements included in our comments. However, if the other improvements are not

implemented we believe application of this adjustment would be necessary to accurately project

and pay outlier payments at the 5.1 percent target.” Id. at 015865. The “other improvements”

the commenter suggested included a proposal for calculating “actual outlier payment percentages

based on actual historical payment data,” id. at 015853 (cleaned up), and employing a projection

method a consultant had developed, id. at 015859. The consultant’s principal method involved

changing the projection period for CCRs from the one-year period HHS had been using and

estimating the fixed loss threshold using the most recent CCRs from provider specific files. Id.

at 015859–015863. The consultant also presented “two alternate methods” that “yield[ed] results

very close” to those of the consultant’s principal method: projecting CCRs quarter-by-quarter or

“estimating the rate of change in CCRs” by using “a recent historical industry-wide average as

the projection factor,” the “exact[] approach [HHS] uses to project charge inflation.” Id. at

015863–64.

In the FY 2014 final rule, HHS finally abandoned its complicated method of determining

the CCR adjustment factor by comparing the operating cost and market basket changes (the

method discussed at length in the Court’s preceding procedural analysis) in favor of the simpler

method of “us[ing] historical data to adjust the CCRs.” 78 Fed. Reg. at 50,978. Specifically,

HHS projected CCRs by “compar[ing] the percentage change in the national average case-

54 weighted operating CCR and capital CCR from the December 2011 update of the PSF to the

national average case-weighted operating CCR and capital CCR from the December 2012 update

of the PSF”—a method “consistent with [its] estimation of charge inflation.” Id. at 50,978–79.

In other words, HHS adopted the last of the commenter’s proposals for improving the fixed loss

threshold projection, an alternative method that “result[ed] in a close approximation” to the

commenter’s consultant’s principal method. AR (FY 2014) at 015864. HHS’s adoption of one

of the commenter’s suggested accuracy-improvement options sufficed to respond to the

commenter’s estimate-adjustment method, which it proposed only as a fallback in the event HHS

rejected its other preferred options. See id. at 015865 (“[I]f the other improvements are not

implemented we believe application of [an estimate adjustment factor] would be necessary to

accurately project and pay outlier payments at the 5.1 percent target.” (emphasis added)).

To be sure, in an earlier sentence, the commenter said that HHS should employ an

estimate adjustment factor if it did “not adopt all the suggestions mentioned in our comments to

improve the outlier projections.” Id. (emphasis added). HHS did not adopt all of the

commenter’s myriad suggestions. But this commenter statement was somewhat confusing,

because some of its suggestions were presented as alternatives—it would not necessarily have

made sense for HHS to adopt them all. And the change in the CCR adjustment-factor derivation

was enough of a shift in the way HHS projected fixed loss thresholds that it was reasonable for

HHS to rest upon this change as its response to the commenter’s multi-pronged comment

directed toward improving projection accuracy. Plaintiffs have not engaged with the conditional,

fallback nature of the commenter’s estimate-adjustment-factor suggestion. See Pls.’ Summ. J.

Mem. at 34. Under these circumstances, the Court declines to hold that the FY 2014 rule was

arbitrary and capricious insofar as it did not discuss the estimate adjustment factor.

55 3. HHS’s explanation for refusing to account for reconciliation in the FY 2013 rule was arbitrary and capricious, but HHS sufficiently explained this decision in the FY 2014–2016 rules.

Plaintiffs next claim that the FY 2013–2016 rules were arbitrary and capricious because

HHS refused to account for the effects of its payment reconciliation process on its projections of

the fixed loss threshold. Pls.’ Summ. J. Mem. at 36. A bit of background is necessary to

understand this set of challenges, for the Court does not write on a clean slate with respect to

these claims. First, a refresher on HHS’s reconciliation process, which HHS has used since 2003

to fight the practice of turbo-charging: HHS can “recalculate a hospital’s eligibility [for an

outlier payment] using actual cost data at the time of settlement,” as opposed to the older cost

data used to determine the hospital’s initial CCR, and, accordingly, its eligibility for a particular

outlier payment. Billings Clinic, 901 F.3d at 306. If it turns out that the hospital was not in fact

eligible for the outlier payment it received, HHS can “claw-back” the undue amount via the

reconciliation process. Id. Thus, by recovering sums through reconciliation, HHS might end up

ultimately paying less in outlier payments for a given year than initial projections suggested.

Over the years, “commenters [have] repeatedly argued” that HHS should incorporate an

expectation that some payments would be reclaimed through reconciliation into its process for

projecting what fixed loss threshold is appropriate to achieve outlier payments that are 5.1

percent of total payments. Pls.’ Summ. J. Mem. at 35. But HHS repeatedly declined to do so.

In Banner Health, hospitals argued that it was arbitrary and capricious in the FY 2007

rule for HHS not to account for payments that would be reclaimed during reconciliation. 867

F.3d at 1356. HHS had explained its decision not to incorporate reconciliations in part on the

ground that it had implemented policies it believed would put an end to turbo-charging, and

therefore to the need for significant resort to the reconciliation process: “[W]e . . . continue to

believe that, due to the policy implemented in the June 9, 2003 outlier final rule, CCRs will no

56 longer fluctuate significantly and, therefore, few hospitals will actually have these ratios

reconciled upon cost report settlement.” 71 Fed. Reg. at 48,149. HHS further explained that

reconciliation recoupments would be difficult to predict and that its projection model aimed to

account for accurate CCRs, so should reasonably estimate post-reconciliation payments anyway:

In addition, it is difficult to predict which specific hospitals will have CCRs and outlier payments reconciled in their cost reports in any given year. We also noted that reconciliation occurs because hospitals’ actual CCRs for the cost reporting period are different than the interim CCRs used to calculate outlier payments when a bill is processed. Our simulations assume that CCRs accurately measure hospital costs and, therefore, are more indicative of postreconciliation than pre- reconciliation outlier payments.

Id. In Banner Health, hospitals argued that even in spite of the anti turbo-charging reforms it

was arbitrary and capricious for the FY 2005 rule “not to forecast that particular hospitals would

continue collecting outlier payments significantly higher than their actual costs.” 867 F.3d at

1352. Concluding that HHS’s explanation that the 2003 reforms would effectively eliminate

turbo-charging was reasonable, the D.C. Circuit held that the FY 2005 rule was not arbitrary and

capricious. Id. The Court then incorporated this same analysis to reject the hospitals’ similar

challenge to the FY 2007 rule. Id. at 1356.

In FYs 2008–2011, HHS refused to take reconciliation into account, each time giving

essentially the same explanation it had given in the FY 2007 final rule. Pls.’ Summ. Jr. Mem. at

35–36 & 36 n.20 (collecting sources). A group of hospitals again sought to challenge these

refusals in Billings Clinic. As relevant here, the hospitals argued that HHS’s assertions that the

2003 reforms obviated the need for reconciliation were undercut by data showing that

reconciliations continued in spite of the reforms. They further argued that HHS’s claim of

difficulty in predicting which hospitals would be subject to reconciliation was not a good reason

to fail to take account of their important impact and that in any event, what mattered was not

57 which specific hospitals would engage in reconciliation but rather the overall total of reconciled

payments. Final Brief for the Appellants at 38–43, Billings Clinic, 901 F.3d 301 (No. 17-5006).

The D.C. Circuit rejected these challenges in Billings Clinic. It held that the Banner

Health court’s conclusion that HHS “‘was under no obligation’ to ‘account for the possibility of

reconciliation in setting the fixed-loss threshold’ . . . applie[d] with equal force to the 2008

through 2011 outlier thresholds.” Billings Clinic, 901 F.3d at 313 (quoting Banner Health, 867

F.3d at 1356). Nothing in the Billings Clinic record, the court held, supported a departure from

the holding that HHS “reasonably concluded ‘that [the] charging practices would not fluctuate

significantly enough to justify accounting for reconciliation[.]’” Id. (quoting Banner Health, 867

F.3d at 1352).

Accordingly, it is the law of the D.C. Circuit that HHS’s decisions not to account for

reconciliation in the FY 2007–2011 rules, based on the explanations repeated across those

rulemakings, were not arbitrary and capricious even in the face of the arguments hospitals made

in Billings Clinic. In refusing to account for reconciliation in the FY 2013–2016 rules, as

relevant here, HHS both repeated the explanations it had used since 2007 (that the 2003 reforms

had reduced the volume of reconciled payments, that there was difficulty in predicting which

hospitals would be subject to reconciliation, and that the model already used accurate CCRs),

and added additional justifications in response to specific comments received for FYs 2014–2016

(discussed below). 77 Fed. Reg. at 53,695 (FY 2013 Rule); 78 Fed. Reg. at 50,979–80 (FY 2014

Rule); 79 Fed. Reg. 49,854. 50,375–78 (Aug. 22, 2014) (FY 2015 Rule); Fed. Reg.49,426,

49,781 (Aug. 17, 2015) (FY 2016 Rule). To succeed on their claims that it was arbitrary and

capricious for HHS not to account for reconciliation in the FY 2013–2016 rules, Plaintiffs must

demonstrate some difference between their challenges and the ones rejected in Billings Clinic.

58 They must point to evidence in the records for the FY 2013–2016 rules that casts new doubt

upon HHS’s previous explanations (repeated in support of the FY 2013–2016 rules), and/or show

that HHS’s new justifications were arbitrary. See Billings Clinic, 901 F.3d at 313 (holding that

“[n]othing in the current record” supported a departure from Banner Health’s decision to uphold

the 2005 and 2007 refusals). Plaintiffs have done so for the FY 2013 rule, but not for the others.

Plaintiffs’ general reconciliation arguments. Plaintiffs take HHS’s repeated justifications

one at a time. Their first attack, on HHS’s explanation that “it is difficult to predict the specific

hospitals that will have CCRs and outlier payments reconciled in any given year,” e.g., 77 Fed.

Reg. at 53,695, comes on two fronts. First, Plaintiffs argue that it is “irrational” to focus on

identification of specific hospitals when what matters is “what the overall total amount of

reconciliation will be.” Pls.’ Summ. J. Mem. at 37. But the Billings Clinic challengers raised

this exact argument in challenging the FY 2008–2011 rules, and the D.C. Circuit rejected it

(albeit without specificity): this argument was part of the “record” the Billings Clinic court held

did not “support[] a different answer” than Banner Health’s decision to uphold the FY 2007 rule,

Billings Clinic, 901 F.3d at 313, which had included the same HHS explanation, 71 Fed. Reg. at

48,149. See Final Brief for the Appellants at 42, Billings Clinic, 901 F.3d 301 (No. 17-5006)

(“HHS’s justification is not logical. To account for reconciliation in setting the threshold, what

mattered was the total amount of reconciled payments, not the identities of the hospitals with

payments at issue. That it was difficult to say which hospitals might undergo reconciliation does

not imply that HHS could not estimate their impact.”). Plaintiffs have not pointed to any

differences in the rulemaking records for the FY 2013–2016 rules that would make this argument

more persuasive in opposition to these rules than it was in opposition to the FY 2008–2011 rules.

Moreover, they have not explained why it was irrational for HHS to be concerned about the

59 feasibility of identifying which hospitals would be subject to reconciliation as a means of

projecting the total amount of reconciled payments—a plausible reading of HHS’s explanations.

See Rural Cellular Ass’n v. FCC, 588 F.3d 1095, 1105 (D.C. Cir. 2009) (“[W]hen an agency’s

decision is primarily predictive, our role is limited; we require only that the agency acknowledge

factual uncertainties and identify the considerations it found persuasive.”).

In their second challenge to this first explanation (the difficulty in identifying hospitals

subject to reconciliation), Plaintiffs shift toward accepting it on its terms, but argue that specific

items in comments from 2013–2016 showed that HHS could in fact identify specific hospitals.

During the FY 2013–2016 rulemakings, a commenter reviewed hospital cost reports and

presented tables showing the total amount of outlier reconciliation payments for years 2003–

2010. AR (FY 2013) at 13103; AR (FY 2014) at 15546; AR (FY 2015) at 18545; AR (FY 2016)

at 20569. But Plaintiffs have not explained why these evaluations of past reconciled payments

(which do not even mention specific hospitals, though presumably are drawn from hospital-

specific data), responded to HHS’s concern about the difficulty in predicting specific hospitals

that would be subject to future reconciled payments. In FY 2013, the commenter also said,

without citation and rather generally, that it was “aware of hospitals that have received

reconciliation notices for significant amounts that are not reflected in” the table submitted for

that year. AR (FY 2013) at 13103. This vague assertion hardly pointed HHS toward a viable

method of predicting reconciliation payments.

Plaintiffs characterize HHS’s next repeated explanation as “the statement ‘that few

hospitals will actually have these amounts reconciled.’” Pls.’ Summ. J. Mem. at 36 (quoting 77

Fed. Reg. at 53,692). Zooming out a bit, it becomes clear that this statement was part of HHS’s

contention that the 2003 reforms would sufficiently diminish the need for reconciliation such that

60 reconciliation recoupments would not much affect total payment outlays: “[W]e continue to

believe that, due to the policy implemented in the June 9, 2003 outlier final rule (68 FR 34494),

CCRs will no longer fluctuate significantly and, therefore, few hospitals will actually have these

ratios reconciled upon cost report settlement.” 77 Fed. Reg. at 53,659. The D.C. Circuit upheld

this very same prediction as “not . . . arbitrar[y]” in Banner Health, 867 F.3d at 1351–52, and as

“reasonabl[e]” in Billings Clinic, 901 F.3d at 1313, so Plaintiffs must point to a clear difference

between the records at issue in those cases and the rulemaking records for the FY 2013–2016

rules challenged here in order to succeed in arguing that this explanation is “a non sequitur,”

Pls.’ Summ. J. Mem. at 36.

Their first effort to do so falls flat. Plaintiffs say that what matters is the total amount of

payments that will be recouped via reconciliation, not whether the number of hospitals that

participate in reconciliation is large or small. For support, they note that “[i]n the FY 2004 rule,

HHS identified 50 hospitals that it thought would be subject to reconciliation.” Id. at 36 (citing

68 Fed. Reg. at 45,476). Even though this was only a few hospitals, “HHS accounted for their

reconciliations,” which it explained would improve its estimate of the final amount of outlier

payments. Id. (citing 81 Fed. Reg. 3,727, 3,729 (Jan. 22, 2016) (HHS’s further explanation of

the FY 2004 rule after a remand)). Plaintiffs overlook the fact that the Billings Clinic

challengers also invoked the FY 2004 rule’s accounting for 50 reconciled hospitals, to no avail.

Final Brief for the Appellants at 41, Billings Clinic, 901 F.3d 301 (No. 17-5006) (“In 2003 HHS

overhauled the outlier regulations due to only about twice as many turbo-charging hospitals, and

in the 2004 threshold rulemaking HHS considered the reconciliation of about 50 hospitals to be

critical to the analysis. It is easy to see why it would be. Reconciliation was intended for

hospitals that ‘disproportionately benefited’ from vulnerabilities in the payment regulation so the

61 50 hospitals would likely be among the largest recipients. HHS has not . . . explained why it

departed from its prior practice and ignored reconciliation for the years here at issue.” (citations

omitted)).

Moreover, Plaintiffs have not explained why HHS’s FY 2004 conclusions from the

record before it then should be relevant to evaluating assertions based on the rulemaking records

for FY 2013–2016. Perhaps HHS during those later years lacked the hospital-specific data it had

had in 2004—data from before the recent enactment of the anti-turbo-charging reforms may have

made it easier to identify hospitals especially susceptible to reconciliation in 2004. See 68 Fed.

Reg. at 45,476 (noting that although “it is difficult to project which hospitals will be subject to

reconciliation of their outlier payments using available data,” HHS had been able to identify

about 50 “hospitals that ha[d] been consistently overpaid recently for outliers” (emphasis

added)). None of the comments submitted during the FY 2013–2016 rulemakings suggested a

viable way to identify any particular hospitals that would need to engage in a large volume of

reconciliation during those years.

Still, something important did change between the records at issue in Banner Health and

Billings Clinic and the FY 2013–2016 records at issue here: in each year, a comment presented

data that purported to show that a sizable volume of reconciliation had occurred between 2003

and 2009 and/or 2010 despite the long-running operation of the 2003 anti-turbo-charging

reforms. Generally, the comments each year posited that the average yearly total of reconciled

payments in this period was between about $12.3 million and $13.6 million. AR (FY 2013) at

13103 (showing a 2004–2009 yearly average of $13,570,975); AR (FY 2014) at 15546 (showing

a 2003–2009 yearly average of $12,256,814); AR (FY 2015) at 18545 (showing a 2003–2010

yearly average of $13,616,803); AR (FY 2016) at 20569 (showing a 2003–2010 yearly average

62 of $13,616,803). This data at least facially called into question the core premise of HHS’s long-

running explanation for not accounting for reconciliation payments when projecting the

threshold: that the 2003 reforms would largely eliminate the need for reconciliation. See, e.g., 71

Fed Reg. at 48,149 (FY 2007 rule); 77 Fed. Reg. at 53,695 (FY 2013 rule). Therefore, once

faced with this new data, it became arbitrary for HHS to simply repeat the same assertion it had

used previously without examining how it held up against the new data.

Yet that is exactly what HHS did in the FY 2013 rule. It noted the total volume of

reconciliation payments the commenter had identified for the 2003–2010 period (though not the

yearly average), but then simply rehearsed its old assertion: “[W]e continue to believe that, due

to the policy implemented in the June 9, 2003 outlier final rule (68 FR 34494), CCRs will no

longer fluctuate significantly and, therefore, few hospitals will actually have these ratios

reconciled upon cost report settlement.” 77 Fed. Reg. at 53,695. HHS’s failure to engage with

data apparently showing that reconciliation had continued long past the 2003 reforms is all the

more striking given that the commenter expressly requested that HHS disclose “data showing

that the recoveries obtained during the reconciliation process are immaterial.” AR (FY 2013) at

13103.

In this litigation, the Secretary responds by asserting that the commenters’ yearly

averages did not present a consistent pattern and indeed masked inconsistency in the data: the

commenter’s own 2013 table, for example, showed that annual outlier payments had jumped

from as low as $652,845 in 2003, to $7,317,950 in 2006, all the way up to $32,701,911 in 2004.

AR (FY 2013) at 13102. Because “the totals for each year varied widely,” the comments did not

“demonstrate that making an adjustment for reconciliation was likely to improve the accuracy of

the Secretary’s projections, rather than simply introducing a new source of potential distortion.”

63 Def.’s Summ. J. Mem. at 33–34. But these points cannot save the rule. For one thing, the

observation that reconciliations varied from year to year does not precisely defend the assertion

that reconciliations would generally decline because of the 2003 reforms. More importantly, the

Secretary’s observations about the commenter’s data are entirely post hoc, and therefore cannot

support the FY 2013 rule. HHS should have explained its misgivings on the record. Thus,

HHS’s FY 2013 failure to engage with data that purported to undermine HHS’s conclusion on

what HHS itself evidently considered an important aspect of the problem—whether

reconciliation payments would be negligible going forward because of the 2003 reforms—was

arbitrary and capricious.

A different result obtains for FYs 2014–2016, because in those rules HHS did what it had

failed to do in FY 2013: it explained why it thought the commenter’s average yearly

reconciliation amounts were immaterial to its fixed-loss threshold projections. Each year, HHS

explained that it did not believe that the commenter’s “relatively small annual amount would

have an impact on the outlier threshold because total outlier payments are approximately $4.3

billion.” 78 Fed. Reg. at 50,980 (FY 2014 rule); 79 Fed. Reg. 49,854. 50,377 (FY 2015 rule); 80

Fed. Reg. at 49,781 (FY 2016 rule referring readers to the FY 2015 response). To argue that

these responses were arbitrary, Plaintiffs point out “that $12.3 million, the low end of the

reconciliation that commenters cited, represents 0.29% of $4.3 billion.” Pls.’ Summ. J. Mem. at

39. Plaintiffs claim that it was arbitrary for HHS to consider a 0.29% impact on total outlier

payments immaterial because HHS frequently adjusted for similarly sized forecasted changes in

payments based on the CCR adjustment factor when setting the threshold. Id. (citing a

forecasted CCR decrease of 0.27 percent in FY 2007 and a 0.27 percent increase in FY 2008).

64 This is an apples and oranges comparison. HHS’s goal was to predict outlier payments as

accurately as feasible in order to project the appropriate threshold. When forecasting payment

changes based on CCR adjustment factors, HHS was using the outputs of a carefully calibrated

model it had already developed. HHS had these numbers, and felt confident in their accuracy, so

it made sense to use them, whatever their magnitude. In contrast, HHS’s assertion that 0.27

percent of total payments reconciled was not a large enough impact to adjust for must be

understood together with the further explanation that shortly followed, that it was difficult to

predict reconciliation for a future year because it was difficult to predict which hospitals would

be subject to reconciliation—an explanation that the Court has already held was reasonable. See

78 Fed. Reg. at 50,980 (FY 2014 rule); 79 Fed. Reg. 49,854. 50,377 (FY 2015 rule); 80 Fed.

Reg. at 49,781 (FY 2016 rule referring readers to the FY 2015 response). HHS’s apparent

conclusion that a 0.29 percent impact was not worth incorporating, understood alongside the

accompanying conclusion that HHS did not have a ready means of predicting this figure each

year and thought it would be difficult to develop one, is not inconsistent with its conclusions that

the 0.27 percent outputs of the carefully CCR calibrated model were worth incorporating. Cf.

Hall v. McLaughlin, 864 F.2d 868, 873 (D.C. Cir. 1989) (“[I]f the court itself finds the past

decisions to involve materially different situations, the agency’s burden of explanation about any

alleged ‘departures’ is considerably less.”). In contrast, cases on which Plaintiffs rely to argue

that HHS was required to explain this apparent inconsistency involved either direct reversals of

previous policies or other inapposite fact patterns. United Steel v. Mine Safety & Health Admin.,

925 F.3d 1279, 1284–85 (D.C. Cir. 2019) (agency shift from a policy requiring certain

examination to take place prior to a miner’s shift to a policy permitting a miner to begin working

before the examination had taken place); County of Los Angeles, 192 F.3d at 1022 (HHS had not

65 adequately explained why it had concluded that data was too unreliable to use for one purpose

but then used the same data for another purpose); Dist. Hosp. Partners, L.P., 786 F.3d at 57–60

(HHS had not sufficiently explained why its notice of proposed rulemaking identified 123 turbo-

charging hospitals to account for reconciliation, while its final rule identified only 50 turbo-

charging hospitals when making the same determination). Nor did any commenter alert HHS to

the alleged inconsistency.

The remaining HHS explanation, repeated in some form across FYs 2007–2011 and

2013–2016, related to the fact that reconciliation occurred because the later, updated, “actual

CCRs for the cost reporting period are different than the interim CCRs used to calculate outlier

payments when a bill is processed.” E.g., 77 Fed. Reg. at 53,695 (FY 2013 final rule). HHS

explained that its model sought to forecast CCRs accurately, so its model was closer to the later,

more accurate CCRs that any reconciliation proceeding would ultimately rely on to determine

the extent to which an outlier payment needed to be recouped. In other words, HHS thought its

model already corrected, at least to an extent, for the same imperfections in CCRs that the

reconciliation process sought to correct for, so there was no great need to factor reconciliation

into the model. E.g., id. Admittedly, HHS spelled out its reasoning in the FY 2007 rule more

clearly than it did in the later rules, but the Court understands HHS to have raised substantially

the same point across the FY 2007–2011 rulemakings. Compare, e.g., id.(FY 2013 final rule)

(“Our simulations assume that CCRs accurately measure hospital costs based on information

available to us at the time we set the outlier threshold.”), with 71 Fed. Reg. at 48,149 (FY 2007

final rule) (“Our simulations assume that CCRs accurately measure hospital costs and, therefore,

are more indicative of postreconciliation than pre-reconciliation outlier payments. As a result,

66 we proposed to continue to omit any assumptions about the effects of reconciliation from the

outlier threshold calculation.”).

Plaintiffs say that this explanation is “simply illogical” because HHS’s model drew upon

CCR data from old cost reports (settled before March of the preceding fiscal year), so HHS could

not assume that the CCRs in its model were accurate predicters of post-reconciliation payments,

which relied on later, updated cost reports. Pls.’ Summ. J. Mem. at 37–38. It is not clear to the

Court that Plaintiffs have demonstrated any failure of logic in HHS’s reasoning; as Plaintiffs

acknowledge, HHS did not simply take the March CCRs but employed a model to project them

forward to approximate what they would be in the coming year. See id. But the Secretary does

not defend HHS’s reasoning on this ground. Instead, he says that the accurate CCR forecast

explanation was ancillary to HHS’s “primary” concern that it would be difficult to model

reconciliation payments. Def.’s Summ. J. Mem. at 32–33. In any event, the Court holds that

HHS’s FY 2014–2016 rules find sufficient support in the non-arbitrary explanations regarding

the contextual insignificance of the amounts of reconciliation payments the commenter identified

and regarding the difficulty in identifying hospitals subject to reconciliation. They therefore

stand regardless of the logical strength of the accurate-CCR-forecast explanation.

Plaintiffs’ arguments regarding the 2013 OIG Report. The Court reaches a similar

conclusion with respect to the last of Plaintiffs’ reconciliation-related challenges. During the FY

2015 and FY 2016 comment periods, a commenter drew HHS’s attention to an OIG report from

2013 which indicated that 158 hospitals had received an usually large amount of outlier

payments over the course of 2008–2011. AR (FY 2015) at 18808–10; AR (FY 2016) at 20611–

20614. The commenter suggested that these unusually high payments would “presumably be

recouped” via reconciliation, and that HHS should factor this anticipated reconciliation recovery

67 into its projection of the fixed loss threshold. AR (FY 2015) at 18808–10; see AR (FY 2016) at

20614. Plaintiffs claim that HHS’s refusal to account for reconciliation related to the hospitals

identified in the 2013 OIG report in both the FY 2015 and FY 2016 final rules was arbitrary and

capricious. Pls.’ Summ. J. Mem. at 40; Pls.’ Combination Summ. J. Opp’n & Reply (“[I]n the

FYs 2015 and 2016 rulemakings, HHS irrationally failed to take account of the likelihood there

would be additional reconciliation, given the high-charge claims discussed in the 2013 report

from HHS’s Office of Inspector General.”); see 79 Fed. Reg. 49,854. 50,377–78 (FY 2015 rule);

80 Fed. Reg. at 49,781–82 (FY 2016 rule).

The Court begins its evaluation of this claim where it left off with Plaintiffs’ previous

reconciliation arguments. In addition to the reasons HHS gave for not relying on the OIG report,

the FY 2015 and 2016 rule’s decisions not to account for any reconciliation rested on the same

explanations the Court has just held were not arbitrary: the insignificance of the amounts of

reconciliation payments presented in comments and the difficulty in identifying which hospitals

would be subject to reconciliation. See 79 Fed. Reg. at 50,377 (FY 2015 rule); 80 Fed. Reg. at

49,781 (FY 2016 rule) (referring readers to the FY 2015 explanations). Therefore, Plaintiffs

must show that the OIG report either undermined these explanations as HHS used them in FYs

2015 and 2016 or so altered the record that HHS was required to reach a different result in spite

of these valid explanations. Plaintiffs have not done so.

For one thing, it is not as if the OIG report definitively concluded that the hospitals would

be subject to reconciliation or were engaged in improper charging practices, much less that any

such reconciliation would occur for FYs 2015 and 2016 in particular. As Plaintiffs acknowledge,

“[t]he OIG did not determine that any of the high-outlier hospitals had engaged in misconduct or

identify what recoupments might be warranted.” Pls.’ Summ. J. Mem. at 40 (citing AR (FY

68 2016) at 20663). The OIG report allowed that in at least some cases, high charges could

accurately match high costs of a treatment. See AR (2015 Rule) at 18839. Thus, the OIG report

did not by itself give rise to a compelling need to account for reconciliation in FY 2015 or FY

2016.

For another, though HHS’s responses to the OIG-report comments may not have been

uniformly airtight, Plaintiffs have not met their burden of persuading the court that they were

arbitrary and capricious. First, Plaintiffs take issue with HHS’s FY 2015 statement that any

unduly high charges would be mitigated by the application of CCRs, which would reduce outlier

payments for these hospitals to their proper level. Pls.’ Summ. J. Mem. at 40–41; 79 Fed. Reg.

at 50,377. Plaintiffs say that this assertion contradicts the very premise of the reconciliation

policy, that CCRs might be inaccurate because they might “lag certain charge increases.” Id.

But as the Secretary points out, the possible lag between CCRs and charge increases for which

reconciliation is designed to correct does not mean that a high-charging hospital’s CCR will

necessarily be incorrect, because high charges do not necessarily indicate that a hospital has

suddenly ramped up charges for non-cost-related reasons (as occurs in the turbo-charging

technique reconciliation is designed to stop). “A cost-to-charge ratio would not overstate costs,

for example, if the hospital’s charges were already high at the time the cost-to-charge ratio was

computed, or if the hospital’s charges have increased in step with parallel increases in the

hospital’s costs.” Def.’s Summ. J. Mem. at 37. Again, the OIG report did not accuse the subject

hospitals of turbo-charging or any other form of charge manipulation, and it acknowledged that

at least some of the high charges might have been legitimate. Therefore, HHS’s 2015

expectation that application of CCRs would obviate the need to subject the OIG-report hospitals

to reconciliation was not obviously inconsistent with its reconciliation policy.

69 Second, Plaintiffs argue that it was irrational for HHS to say in the FY 2015 rule that the

hospitals flagged in the OIG report would not end up reconciling because they probably did not

meet HHS’s criteria for automatic reconciliation. See 79 Fed. Reg. at 50,377–78. Plaintiffs say

that this assumption ignored HHS’s policies that a fiscal intermediary could, in its discretion,

subject any hospital to reconciliation, even if the hospital did not automatically qualify, and that

an intermediary could, if it believed a hospital was using an inaccurate CCR, “specify an

alternative CCR.” Pls.’ Summ. J. Mem. at 41 (citation omitted). The FY 2016 explanation even

expressly acknowledged the possibility of specification of alternative CCRs. 80 Fed. Reg. at

49,781; see Pls.’ Summ. J. Mem. at 41–42. But nothing in the OIG report or the associated

comment gave HHS any reason to believe that fiscal intermediaries would apply their

discretionary reconciliation or alternative CCR authority to the hospitals the OIG report flagged.

See Def.’s Summ. J. Mem. at 37.

Third, in FY 2016, the commenter questioned HHS’s FY 2015 assertion that the

application of CCRs would prevent the OIG-report hospitals from receiving undue outlier

payments by referring HHS to the OIG report’s observation that the flagged hospitals had

materially similar CCRs to other hospitals (and had similar lengths of stay)—yet still charged

substantially more for the same procedures. AR (FY 2016) at 20613. HHS responded in the FY

2016 final rule, in part, by noting that the 2008–2011 CCRs used in the OIG report may not have

been fully updated:

The CCRs are updated in the PSF at the time the MAC tentatively settles the hospital cost report, which is approximately 6 to 7 months after the cost report has been submitted. Thus, there is a lag in CCRs with the possibility that a CCR may be 18 months old from the time the cost report is submitted by the provider to the MAC until it is updated at the following tentative settlement. Because hospitals typically increase their charges, over time CCRs will decrease but, due to the lag these lower CCRs will not be reflected in the PSF until the following tentative

70 settlement. Thus, it is possible that the PSF will reflect CCRs that are similar for hospitals with high and low outlier payments.

80 Fed. Reg. at 49,781. Plaintiffs claim that this response “[c]ontradict[ed] HHS’s response

from the previous year,” but they do not explain how. Pls.’ Summ. J. Mem. at 41. This response

was an explanation of HHS’s previous-year statement—that application of CCRs would mitigate

unduly high hospital charges—in response to the 2016 commenter’s challenge to this statement,

not a contradiction of it. Plaintiffs also attack HHS’s response because HHS “did not explain

why . . . high-outlier claims were irrelevant. The time lag in updating CCRs does not prevent

HHS from reconciling a hospital’s outlier payments, it is the whole point of reconciliation.” Id.

But this argument essentially re-hashes Plaintiffs’ objection to the initial 2015 CCR explanation,

which, as the Court has explained, is not clearly inconsistent with reconciliation’s purpose of

preventing hospitals from taking advantage of the time-lag in CCRs.

Fourth, and finally, Plaintiffs point to another part of HHS’s response to the FY 2016

comment. HHS observed that hospitals “determine what they will charge for items, services, and

procedures provided to patients, and these charges are the amount that the [hospitals] bill for an

item, service, or procedure. Moreover, different hospitals can have similar lengths of stay but

different CCRs.” 80 Fed. Reg. at 49,781–82. Plaintiffs complain that this statement ignored that

the OIG report compared hospitals with similar lengths of stay. Pls.’ Summ. J. Mem. at 41. To

the contrary, this statement responded to this feature of the OIG report and explained why HHS

thought the report did not undermine HHS’s conclusion that lower CCRs would reduce outlier

payments to the OIG-report hospitals. Absent any further argument from Plaintiffs on this point,

HHS’s response was not arbitrary. To wrap up, the 2013 OIG report and Plaintiffs’ criticisms of

HHS’s discussion of it do not disturb the Court’s conclusion that HHS acted lawfully when it

71 declined to account for reconciliation when setting the FY 2015 and FY 2016 fixed loss

thresholds.

4. The proper remedy is remand of the FY 2012 and FY 2013 rules for further proceedings consistent with this opinion.

At this point, it makes sense to take stock of where the foregoing discussion leaves us.

The Court has held that none of the challenged rules were procedurally deficient. However, the

FY 2012 and FY 2013 rules were arbitrary and capricious in that HHS did not sufficiently

explain its rejection of the commenter suggestion to use an estimate adjustment factor to improve

the results of its model. The FY 2013 rule suffered from the additional defect of failing to

sufficiently explain HHS’s decision not to account for reconciliation.

Plaintiffs ask the Court to vacate these rules “and remand for recalculation of the

hospitals’ outlier payments.” Pls.’ Summ. J. Mem. at 42. The Secretary contends that the Court

should instead simply remand to the Secretary for further action consistent with this opinion.

Def.’s Summ. J. Mem. at 39; Def.’s Reply Mem. Supp Cross Mot. Summ. J. at 23. Based on the

nature of the infirmities the Court has found in the FY 2012 and FY 2013 rules, the Court agrees

with the Secretary that remand is the proper course.

It is true that “[t]he ordinary practice . . . is to vacate unlawful agency action.” Standing

Rock Sioux Tribe v. United States Army Corps of Engineers, 985 F.3d 1032, 1050 (D.C. Cir.

2021). But the nature of the agency’s legal violation informs the question of the proper remedy.

When, as here, the agency has failed to sufficiently explain its rules, “the proper course, except

in rare circumstances, is to remand to the agency for additional investigation or explanation.”

Dist. Hosp. Partners, L.P., 786 F.3d at 60 (citation omitted). Before vacatur, “[i]n such

circumstances, the agency must first be afforded an opportunity to articulate, if possible, a better

explanation.” Banner Health, 867 F.3d at 1357 (citation omitted and cleaned up). Thus, when

72 the D.C. Circuit held in Banner Health that the FY 2004, 2005, and 2006 fixed-loss threshold

rules were arbitrary and capricious due to certain failures of explanation, it remanded the rules so

that HHS would have a chance “to remedy the explanatory deficiencies. Id. at 1343 , 1345,

1349–50, 1356–7; see also Dist. Hosp. Partners, LP, 786 F.3d at 63 (holding that “the 2004

outlier threshold . . . [was] inadequately explained and” instructing the district court to “remand

the 2004 rule to the Secretary for further proceedings consistent with this opinion”); County of

Los Angeles, 192 F.3d at 1023 (“While we have identified significant inconsistencies and gaps in

the Secretary’s rationale for using the 1981 MEDPAR file [to set outlier thresholds for fiscal

years 1985–1986], bedrock principles of administrative law preclude us from declaring

definitively that her decision was arbitrary and capricious without first affording her an

opportunity to articulate, if possible, a better explanation.”).

The Court follows Banner Health, District Hospital Partners, and County of Los Angeles

because they are directly on point: each remanded without vacating when it found that annual

fixed-loss threshold rules were arbitrary and capricious because they suffered from explanatory

failures. For good measure, the Court notes that the Allied Signal factors, which the D.C. Circuit

has applied to guide the vacatur-or-remand decision in many cases but not in fixed-loss threshold

cases, support the decision to remand. Under this test, “[t]o determine whether to remand

without vacatur, [courts] consider[] first, the seriousness of the action’s deficiencies, and,

second, the likely disruptive consequences of vacatur.” Am. Great Lakes Ports Ass’n v. Schultz,

962 F.3d 510, 518–19 (D.C. Cir. 2020) (cleaned up) (citing Allied-Signal, Inc. v. Nuclear Regul.

Comm’n, 988 F.2d 146, 150–51 (D.C. Cir. 1993)).

Plaintiffs offer only two reasons in support of their argument that the seriousness-of-

deficiencies factor supports vacatur. This factor “is determined at least in part by whether there

73 is ‘a significant possibility that the [agency] may find an adequate explanation for its actions’ on

remand.” Standing Rock Sioux Tribe, 985 F.3d at 1051. First, Plaintiffs say that HHS’s shift to a

simplified CCR-adjustment-factor method in FY 2014 would preclude HHS from defending its

previous method on remand. Pls.’ Summ. J. Mem. at 43. But the first specific violation the

Court has identified is not HHS’s decision to employ its complex, pre-2014 method in the FY

2012 and 2013 rules. Rather, the Court has held that HHS did not provide a sufficient

explanation during these years of its rejection of the proposal to use an estimate adjustment

factor alongside, and as a corrective to, the complex method. HHS’s adoption of a simplified

method for FY 2014 does not necessarily preclude it from coming up with an adequate

explanation for its decision not to add an estimate adjustment factor to the complex method it

used in FY 2012 and FY 2013. Plaintiffs do not argue that HHS’s FY 2014 switch to a

simplified CCR-adjustment method precludes HHS from adequately supplementing on remand

the second explanation the Court has found lacking, regarding HHS’s decision not to account for

reconciliation in the FY 2013 rule. Indeed, the decision whether to account for reconciliation

does not appear to turn on HHS’s choice of CCR-adjustment method—Plaintiffs have challenged

HHS’s decisions not to account for reconciliation both before and after the FY 2014 switch to a

simplified method.

Next, Plaintiffs say that because HHS eventually determined it could (and would)

account for reconciliation in FY 2020, HHS could not conceivably explain on remand its

decision not to do so in FY 2013. See id. But there may have been circumstances in 2013 that

made it more difficult to predict reconciliation at that time than it was in 2020. Moreover, the

specific defect the Court has identified was HHS’s FY 2013 failure to engage with commenter

74 data tending to show that a certain amount of reconciliation was taking place. HHS may be able

to sufficiently support its decision with further explanation, as it did in the FY 2014–2016 rules.

As for the consequences of vacatur, Plaintiffs identify the following: “[i]f HHS does, as

the Hospitals predict, end up lowering the threshold, hospitals that have available procedural

mechanisms may be able to seek additional outlier payments.” Id. at 44. Plaintiffs say that this

would not be disruptive because it would not require broader corrective action beyond

recalculating specific hospital cost reports. See id. But the D.C. Circuit has held that remand

without vacatur “is appropriate when vacatur would disrupt settled transactions.” Am. Great

Lakes Ports Ass’n, 962 F.3d at 519 (upholding the district court’s decision to remand without

vacating when vacating the challenged rate-setting rule would mean that past payments were

erroneous). Because any necessary adjustment of outlier payments to Plaintiffs for FYs 2012

and 2013 could wait until after the Secretary has had a chance to decide whether to revise its

responses on remand or proceed otherwise, the Court holds that this factor weighs in favor of

remanding. Therefore, the Court remands the FY 2012 and FY 2013 rules, without vacatur, for

further proceedings consistent with this opinion.

Given this disposition, the Court finds it unnecessary at this time to reach Plaintiffs’

remaining claim, that it was arbitrary and capricious in the FY 2012 and FY 2013 rules to

continue using the same CCR adjustment method even in the face of evidence that, according to

Plaintiffs, demonstrated that HHS had routinely underestimated the decline in CCRs and that it

had frequently failed to achieve its 5.1 percent payment target. See Pls.’ Summ. J. Mem. at 26–

34. Though the parties dispute the strength of the evidence supporting these asserted trends, see

Def.’s Summ. J. Mem. at 25–27, Plaintiffs’ argument on this claim is closely related to the trend

data the Court has found concerning enough to require a persuasive HHS response in the context

75 of the estimate adjustment factor comments. See supra at 50. Thus, the Court’s remand

regarding the estimate adjustment factor will require HHS to reconsider its fixed-loss threshold

projection for FYs 2012 to 2013, including in a way that relates closely to Plaintiffs’ remaining

challenge. Or, on remand, the Secretary may choose to apply a new approach to the FY 2012

and 2013 thresholds rather than defend its original method. See County of Los Angeles, 192 F.3d

at 1023 (noting that on remand, the Secretary could “either . . . recalculate outlier thresholds for

fiscal years 1985–1986 or . . . offer a reasonable explanation for refusing to use [certain data] in

setting outlier thresholds during those years”).

Because it is unclear whether Plaintiffs’ contrary-to-trend argument will take the same

shape, or be relevant at all, after remand, the Court declines to adjudicate it now. See Pub.

Citizen v. Fed. Motor Carrier Safety Admin., 374 F.3d 1209, 1216 (D.C. Cir. 2004) (holding that

a rule was arbitrary and capricious in one respect but not reaching other claims that “also raise[d]

troubling concerns about the decisionmaking process” because on remand “the agency [would]

be free in its further proceedings to consider the other objections anew in light of this opinion

and its own responses” to the issue that was the basis for the remand); cf. BP Energy Co. v.

FERC, 828 F.3d 959, 969 (D.C. Cir. 2016); Nassar & Co. v. SEC, 566 F.2d 790, 794 (D.C. Cir.

1977); Exxon Co., U.S.A. v. FERC, 182 F.3d 30, 45 (D.C. Cir. 1999); Oceana, Inc. v. Ross, 363

F. Supp. 3d 67, 70 (D.D.C. 2019). If HHS does choose to maintain its original course on

remand, it would do well to seriously engage with Plaintiffs’ complaints about the underpayment

trends they claim to have identified—Billings Clinic, 901 F.3d at 315 provides support for the

general proposition that HHS was not entitled to rely on the same method and explanations in the

face of any sustained pattern of underpayments.

76 5. The Court will not revisit its previous holding that certain of Plaintiffs’ claims are timely.

The Court’s decision to remand on the ground that the FY 2012 and FY 2013 rules were

arbitrary and capricious in the ways the Court has described requires it to resolve one additional

issue. The Secretary asks the Court to reconsider its earlier ruling that certain plaintiffs’ claims

relating to (among other rules) the FY 2012 and FY 2013 rules were timely even though they

were added to the action by amended complaint rather than supplemental complaint. Under 42

U.S.C. § 1395oo(f)(1), a party must bring any district court action pertaining to a PRRB judicial-

review grant within 60 days of the decision. The plaintiffs added the claims in question within

60 days by amended complaint. Because a party may not add new claims based on events that

took place after the initial filing of the suit by amended complaint as of right, and instead must

do so by obtaining leave of court to file a supplemental complaint, the Secretary argues that these

claims were not filed within the 60-day period. As the Court summarized previously:

The Secretary suggests that, because these particular claims had been approved for expedited judicial review after the filing of the original complaints, they should have been added by supplementation under Rule 15(d) (which requires court approval). As a result, according to the Secretary, these claims were filed “without legal effect” and are now barred, because they were not actually submitted within 60-day period for filing a district court action under § 1395oo(f)(1).

Mot. Dismiss Op. at 26–27 (citations omitted). In response to the Secretary’s motion to dismiss

on this ground, the Court agreed that plaintiffs were required to add the in-question claims by

supplemental complaint rather than amended complaint, but nevertheless “exercise[d] its

discretion to treat the mislabeled amended complaint as a supplemental complaint, nunc pro

tunc.” Id. at 30. The Court explained that the Secretary had not identified any prejudice due to

the hospitals’ mislabeling of what was in effect a supplemental complaint as an amended

complaint, that the Secretary could have moved to strike the filing but instead waited until after

the 60-day period had passed before bringing the mislabeling issue to the Court’s attention, and

77 that there was no reason to think the Court would not have granted leave to file a supplemental

complaint if the hospitals had requested it at the proper time. Id. at 29–30.

Now, the Secretary asks the Court to “reconsider[]” this decision. Def.’s Suppl. Mem. at

41. Reconsideration of an interlocutory order may be appropriate “when a court has ‘patently

misunderstood the parties, made a decision beyond the adversarial issues presented, made an

error in failing to consider controlling decisions or data, or where a controlling or significant

change in the law has occurred.’” Ali v. Carnegie Inst. of Wash., 309 F.R.D. 77, 80 (D.D.C.

2015) (quoting U.S. ex rel. Westrick v. Second Chance Body Armor, Inc., 893 F. Supp. 2d 258,

268 (D.D.C. 2012)). A court’s failure to consider “controlling decisions . . . that might

reasonably be expected to alter [its] conclusion” may also be grounds for reconsideration.

Jordan v. U.S. Dep’t of Lab., 308 F. Supp. 3d 24, 35 (D.D.C. 2018) (citation omitted).

The Secretary’s principal contention is that the Court’s decision was contrary to Supreme

Court precedent holding “that a court cannot issue a nunc pro tunc order to revise the date of a

court action or otherwise alter the history of a case.” Def.’s Summ. J. Mem. at 41–42. For this,

the Secretary relies on Roman Cath. Archdiocese v. Acevedo Feliciano, 140 S. Ct. 696, 700–01

(2020) (per curiam). But the Secretary reads this case too broadly. In Acevedo Feliciano, a case

had been removed from the Puerto Rico Court of First Instance to federal district court. Id. at

700. The district court granted a motion to remand on August 20, 2018, but attempted to do so

nunc pro tunc on to May 13, 2018, the day it had likely lost subject matter jurisdiction on

related-to-bankruptcy grounds due to a bankruptcy court’s dismissal of a related bankruptcy

proceeding. Id. at 701. The Supreme Court held that orders the Court of First Instance issued

between May 13 and August 20 were void, because the Court of First Instance lacked any

jurisdiction until the case had been remanded. Id. at 700. It did not matter that the district court

78 had attempted to remand nunc pro tunc as of May 13. This attempted relation back was not

effective, because “nothing occurred in the [d]istrict [c]ourt on March 23, 2018.” Id. at 701. The

Supreme Court explained that a nunc pro tunc order may be issued only “to reflect the reality of

what has already occurred. . . . Such a decree presupposes a decree allowed, or ordered, not

entered, through inadvertence of the court.” Id. at 700–01. The purpose of a nunc pro tunc

decree “is to correct mistakes or omissions in the record so that the record properly reflects the

events that actually took place.” Rohe v. Wells Fargo Bank, N.A., 988 F.3d 1256, 1261 n.6 (11th

Cir. 2021) (citation and emphasis omitted).

Unlike in Acevedo Feliciano, in which the district court sought to relate its nunc pro tunc

decree to an event (the bankruptcy court dismissal) that occurred in another court, in this case,

the Court’s nunc pro tunc granting of leave to file a supplemental complaint did relate to

something that had occurred in the district court: the filing of the amended complaints that

should have been labeled as supplemental complaints. As the Court explained when it resolved

the motion to dismiss, to the extent the “amended complaints” added claims based on PRRB

expedited review grants that took place after the initial filing of the action, they were, by

definition, not actually “amended complaints”—despite being labeled that way. Instead, they

were properly understood as supplemental complaints. Mot. Dismiss. Op. at 27 (“The difference

between supplementation and amendment is that an amendment ‘typically rest[s] on matters in

place prior to the filing of the original pleading,’ while a supplement ‘sets forth transactions or

occurrences or events which have happened since the date of the pleading sought to be

supplemented.” (quoting United States v. Hicks, 283 F.3d 380, 385 (D.C. Cir. 2002) (internal

quotation marks omitted)). The “amended complaints” were simply “mislabel[ed].” Id. at 30.

Thus, by granting leave to add these claims nunc pro tunc, the Court was simply acting “to

79 reflect the reality of what ha[d] already occurred,” Acevedo Feliciano, 140 S. Ct. 696, 700–01

(cleaned up): the filing of a supplemental complaint. The Court did not backdate any new filing

or order; it simply “treat[ed] the mislabeled amended complaint as a supplemental complaint,

nunc pro tunc.” Mot. Dismiss Op. at 30.15 Given these differences between Acevedo Feliciano

and the instant case, Acevedo Feliciano is not a controlling decision that makes reconsideration

appropriate.16

Beyond asserted inconsistency with Supreme Court precedent, the Secretary says that the

Court’s motion-to-dismiss decision was outside the adversarial issues presented, but does not

explain how. Def.’s Summ. J. Mem. at 40. In any event, the Court does not find any of the

Secretary’s remaining arguments to justify reconsideration of its initial holding. The Secretary

takes issue with the Court’s previous conclusion that the timing of his timeliness argument had a

15 The fact that Plaintiffs did not move for leave to file a supplemental complaint when they filed their “amended complaints” does not alter the Court’s conclusion. Had the filings been properly labeled as a supplemental complaint, the Court would have had the discretion to treat the supplemental complaint as including the required request to file one. See In re Regions Morgan Keegan Sec., Derivative v. Morgan Asset Mgmt., Inc., No. 08-2260, 2010 WL 11441471, at *1 (W.D. Tenn. Jan. 4, 2010) (finding that a non-pro se plaintiff’s filing, though labeled “Amended Complaint,” was actually a supplemental complaint and construing it as a motion to file a supplemental complaint); cf. Mangwiro v. Napolitano, 939 F. Supp. 2d 639, 647 (N.D. Tex. 2013) (construing a supplemental complaint “as a motion to supplement Plaintiff’s Complaint because, although not styled as a motion, it includes a request by Plaintiffs to supplement their pleadings”), aff’d sub nom. Mangwiro v. Johnson, 554 F. App’x 255 (5th Cir. 2014); Costa v. Bazron, No. CV 19-3185, 2020 WL 1935524, at *1 (D.D.C. Apr. 22, 2020) (“Although Plaintiffs seek leave to amend their complaint under Rule 15(a)(2), the Court will construe the pleading as a motion to file a supplemental complaint pursuant to Rule 15(d).” (citation omitted)). As the Court explained when denying the motion to dismiss, “there is no reason to think the Court would not have granted the motion to supplement, given the interests in judicial economy,” Mot. Dismiss Op. at 29–30, and even now the Secretary has not offered any reason why such a motion should not have been granted. 16 The Secretary also cites a Supreme Court case he says indicates that the § 1395oo(f)(1 60-day period is not subject to equitable tolling, Def.’s Summ. J. Mem. at 42 (quoting Sebelius v. Auburn Reg’l Med. Ctr., 568 U.S. 145, 159–60 (2013), but the Court’s decision that the supplementary claims are not untimely was not and is not based on equitable tolling.

80 “‘gotcha’ quality,” Mot. Dismiss Op. at 29, insisting that he warned Plaintiffs’ counsel about the

interaction between the requirement of court leave to file a supplemental complaint and the 60-

day § 1395oo(f)(1) period during the Banner Health litigation. But the Court will not rest its

decision on events that took place outside the record of this case. And in any case, the Court’s

initial point remains: the Secretary could have raised this issue earlier by moving to strike the

improper amended complaint. See Mot. Dismiss Op. at 29. Moreover, the Court notes that the

Secretary’s summary judgment brief does not call into question a key premise of the Court’s

motion-to-dismiss ruling, that “the Secretary has not pointed to any undue prejudice that has

resulted from the mislabeling.” Id. at 30. The Court holds that reconsideration of is motion-to-

dismiss holding is not warranted. Therefore, the supplemental claims are timely.

VI. CONCLUSION

For the foregoing reasons, Plaintiffs’ Motion to Complete Administrative Records (ECF

No. 184) is GRANTED IN PART AND DENIED IN PART; Plaintiffs’ Motion for Summary

Judgment (ECF No. 185) is GRANTED IN PART AND DENIED IN PART; and Defendant’s

Cross-Motion for Summary Judgment (ECF No. 188) is GRANTED IN PART AND DENIED

IN PART. An order consistent with this Memorandum Opinion is separately and

contemporaneously issued.

Dated: 06/17/2022 RUDOLPH CONTRERAS United States District Judge